Intelligent detection method and system for cross-sectional characteristics of a composite bar wire round hole pass

Abstract

The application discloses a kind of composite bar wire round hole type material cross section feature intelligent detection method and system, the detection method includes calculating first image pixel scale according to calibration plate image;Edge image is obtained according to material cross section image;The circumference length of each pixel point of outer edge is calculated;According to the leftmost pixel point and the rightmost pixel point of first image pixel scale and outer edge horizontal, the circumference width of material and its endpoint coordinates are calculated;Material height is calculated according to the endpoint coordinates of material circumference width;The coating thickness of each pixel point of core edge is calculated according to edge image and material circumference width;Core detection area and total detection area are calculated, and the coating detection area is calculated according to the core detection area and total detection area.The application realizes the intelligent, high-precision detection of multiple characteristic data of composite bar wire round hole cross section, and provides accurate data support for process and pass optimization of composite bar wire.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of workpiece feature detection, and particularly relates to a composite bar wire round pass blank cross section feature intelligent detection method and system. BACKGROUND

[0002] The stainless steel-carbon steel composite bar wire is a kind of composite steel material with an outer layer of stainless steel and a core of carbon steel. The outer layer of stainless steel has good corrosion resistance, and the price is about one third of that of stainless steel, which has significant advantages. The preparation process of the stainless steel-carbon steel composite bar wire includes non-oxidizing assembly of stainless steel pipe and carbon steel round bar, rolling forming of the blank, pickling, etc. This process has initially realized the industrialized trial production of stainless steel composite reinforcing bar, stainless steel composite sucker rod and stainless steel cladding grounding rod, etc.

[0003] During the rolling process of the stainless steel-carbon steel composite bar wire, there are differences in mechanical properties between the double metals compared with the traditional single metal bar wire rolling, which leads to differences in elongation of the blank, deflection, uneven distribution of the stainless steel cladding, etc. Generally, the head and tail samples cut off by the flying shear machine during the rolling process of the composite bar wire are round passes. Taking the existing thread rolling of a certain steel plant as an example, different sizes of thread steel have different rolling passes, and the rolling mill is distributed horizontally and vertically. The rolling is usually divided into rough, medium and (pre) finishing rolling. For example, the rolling of Φ28mm straight thread steel has a total of 14 passes, and the rough, medium and finishing rolling has 6-4-4 passes respectively, with a total of 2 flying shear machines. The rolling of Φ12mm disc thread steel has a total of 22 passes, and the rough, medium, pre-finish and finish rolling has 6-6-6-4 passes respectively, with a total of 3 flying shear machines, and the flying shear machines obtain round pass blanks.

[0004] For the blanks obtained by the flying shear machine, emergency stop or rolling mill steel stacking during the rolling process of the composite bar wire, the description of the features (cladding area, thickness distribution of the stainless steel cladding, etc.) of the blank by the on-site technical personnel through the naked eye is qualitative. Although subtle differences can be distinguished, it is difficult to determine the specific magnitude, and it is easy to be affected by factors such as light, experience, subjective psychology and vision, making it difficult to achieve accurate, standard and objective evaluation of the features. SUMMARY

[0005] The purpose of the present application is to provide a composite bar wire round pass blank cross section feature intelligent detection method and system to solve the problem that the traditional technology is difficult to achieve accurate, standard and objective evaluation of the blank features by qualitatively describing the features of the blank with the naked eye.

[0006] The present application solves the above technical problems by the following technical solutions: a composite bar wire round pass blank cross section feature intelligent detection method, comprising:

[0007] obtaining a calibration plate image and a blank cross section image;

[0008] calculating a first image pixel ruler according to the calibration plate image;

[0009] obtaining an edge image according to the cross-sectional image of the material form, wherein the edge image comprises an outer layer edge and a core edge;

[0010] calculating a material form circumferential length of each pixel point of the outer layer edge according to the edge image;

[0011] calculating a material form circumferential width and its endpoint coordinates according to the first image pixel ruler and the leftmost and rightmost horizontal pixel points of the outer layer edge;

[0012] calculating a material form height according to the endpoint coordinates of the material form circumferential width;

[0013] calculating a coating thickness of each pixel point of the core edge according to the edge image and the material form circumferential width;

[0014] calculating a core detection area and a total detection area according to the edge image, and calculating a coating detection area according to the core detection area and the total detection area.

[0015] Further, the acquisition process of the calibration plate image and the cross-sectional image of the material form is as follows:

[0016] acquiring a cross-sectional sample of the material form;

[0017] polishing the cross-sectional sample of the material form using a polishing machine, etching the polished cross-sectional sample of the material form using a 4% nitric acid alcohol etching solution, and then cleaning and drying to obtain a sample;

[0018] placing a calibration plate on one side of the sample, keeping the width direction of the sample horizontal, and collecting calibration plate and sample images;

[0019] cropping and segmenting the calibration plate and sample images to obtain a calibration plate image and a cross-sectional image of the material form.

[0020] Further, calculating a first image pixel ruler according to the calibration plate image comprises:

[0021] preprocessing the calibration plate image; wherein the calibration plate image comprises a plurality of calibration circles;

[0022] extracting a plurality of connected regions from the preprocessed calibration plate image, and measuring the area of each connected region and the major axis length and minor axis length of an ellipse having the same normalized second central moment as each connected region;

[0023] extracting a connected region i that satisfies both condition one and condition two:

[0024] Condition one: ;

[0025] Condition two: ;

[0026] wherein, , respectively represent the area of the multiple connected region i, j; represents the area accuracy threshold value; represents the number of multiple connected regions extracted from the pre-processed calibration plate image; , respectively represent the major axis length and the minor axis length of the ellipse with the same normalized second central moment as the multiple connected region i;

[0027] According to the extracted multiple connected regions that simultaneously satisfy condition one and condition two, the first image pixel ruler is calculated, and the specific calculation formula is:

[0028] ;

[0029] wherein, represents the first image pixel ruler; represents the diameter of the calibration circle in the calibration plate; n represents the number of extracted multiple connected regions that simultaneously satisfy condition one and condition two.

[0030] Further, the material type circumference length of each pixel point of the outer layer edge is calculated according to the edge image, including:

[0031] Taking the leftmost horizontal pixel point of the outer layer edge as the first point, all the pixel points of the outer layer edge are sorted, and the sequence number of the highest vertical pixel point is recorded as Nb', the sequence number of the rightmost horizontal pixel point is recorded as Nc', and the sequence number of the lowest vertical pixel point is recorded as Nd';

[0032] The distance between each pixel point of the outer layer edge and other pixel points is calculated, and the specific calculation formula is:

[0033] ;

[0034] wherein, represents the distance between the i-th pixel point and the j-th pixel point of the outer layer edge, , , represents the number of pixel points of the outer layer edge; represents the coordinates of the i-th pixel point of the outer layer edge; represents the coordinates of the j-th pixel point of the outer layer edge;

[0035] find the maximum distance between each pixel point of the outer edge and other pixel points, and determine whether the maximum distance is greater than a first reference distance, if not, the first reference distance is the maximum distance; wherein the determination of the first reference distance is as follows:

[0036] If all pixel points of the outer edge are sorted in a clockwise direction, and i

[0037] If all pixel points of the outer edge are sorted in a counterclockwise direction, and i

[0038] The maximum distance between each pixel point of the outer edge and other pixel points is the length of the circumference of the material type of each pixel point of the outer edge.

[0039] Further, the length of the circumference of the material type and the coordinates of its endpoints are calculated according to the first image pixel scale and the leftmost and rightmost pixel points of the outer edge, including:

[0040] Take the leftmost pixel point of the outer edge as the first point, sort all pixel points of the outer edge, and record the serial number of the rightmost pixel point as Nc';

[0041] Determine the line between the leftmost and rightmost pixel points of the outer edge;

[0042] According to the line, determine the sequence number deviation between the endpoint serial numbers of the length of the circumference of the material type and the leftmost and rightmost pixel points of the outer edge, and further determine the endpoint serial numbers of the length of the circumference of the material type;

[0043] According to the endpoint coordinates of the length of the circumference of the material type and the first image pixel scale, calculate the length of the circumference of the material type, and the specific calculation formula is:

[0044] ;

[0045] wherein, represents the length of the circumference of the material type; , respectively represent the coordinates of the two endpoints of the length of the circumference of the material type; represents the first image pixel scale.

[0046] Further, the detection method further comprises correcting the circumferential width of the material type, specifically comprising:

[0047] obtaining the actual width of the material type cross section, and calculating the width error rate according to the actual width of the material type cross section and the circumferential width of the material type;

[0048] judging whether the width error rate is less than the error rate threshold value, and if not, calculating the second image pixel ruler according to the end point coordinates of the circumferential width of the material type and the actual width of the material type cross section, and the specific calculation formula is:

[0049] ;

[0050] wherein, represents the second image pixel ruler; represents the actual width of the material type cross section; , respectively represents the coordinates of the two end points of the circumferential width of the material type;

[0051] calculating the corrected circumferential width of the material type according to the second image pixel ruler and the end point coordinates of the circumferential width of the material type, and the specific formula is:

[0052] ;

[0053] wherein, represents the corrected circumferential width of the material type.

[0054] Further, the height of the material type is calculated according to the end point coordinates of the circumferential width of the material type, specifically comprising:

[0055] determining the horizontal coordinate of the midpoint of the width direction according to the end point coordinates of the circumferential width of the material type;

[0056] determining the horizontal coordinate value range according to the set horizontal coordinate deviation and the horizontal coordinate of the midpoint;

[0057] determining the upper intersection point sequence number and the lower intersection point sequence number of the vertical line passing through each horizontal coordinate in the horizontal coordinate value range and the outer edge, and further determining the corresponding vertical coordinate and the midpoint of the width direction of each horizontal coordinate in the horizontal coordinate value range;

[0058] determining the upper groove bottom arc center point coordinates corresponding to each midpoint according to the vertical distance difference between the upper groove bottom arc center point of the roll pass and the actual midpoint of the width direction, and determining the lower groove bottom arc center point coordinates corresponding to each midpoint according to the vertical distance difference between the lower groove bottom arc center point of the roll pass and the actual midpoint of the width direction;

[0059] setting the value range of the upper intersection point sequence number and the lower intersection point sequence number;

[0060] calculate distances between each upper intersection point in the value range of the upper intersection point sequence number and the corresponding upper groove bottom arc center point, and distances between each lower intersection point in the value range of the lower intersection point sequence number and the corresponding lower groove bottom arc center point, to form a distance set composed of all the distances;

[0061] calculate variances of each distance in the distance set;

[0062] select the intermediate point corresponding to the distance with the smallest variance as the profile center point;

[0063] determine the upper intersection point and the lower intersection point of the vertical line passing through the profile center point and the outer edge, and calculate the profile height according to the upper intersection point and the lower intersection point.

[0064] Further, according to the edge image and the profile circumferential spread, the coating thickness of each pixel point of the core edge is calculated, comprising:

[0065] take the left intersection point of the straight line where the profile circumferential spread is located and the core edge and the outer edge as the first point of the core edge and the outer edge, respectively, and sort all pixel points of the core edge and the outer edge;

[0066] take the distance between the first point of the core edge and the first point of the outer edge as the second reference distance;

[0067] calculate the distance between each pixel point of the core edge and other pixel points of the outer edge; wherein the other pixel points of the outer edge refer to the pixel points other than the first pixel point of the outer edge;

[0068] find the minimum distance between each pixel point of the core edge and other pixel points of the outer edge;

[0069] if the minimum distance is less than the second reference distance, the minimum distance is the coating thickness of the corresponding pixel point of the core edge; if the minimum distance is greater than or equal to the second reference distance, the second reference distance is the coating thickness of the corresponding pixel point of the core edge.

[0070] Further, the detection method further comprises calculating a profile circumferential evaluation index to further evaluate the profile circumferential characteristics; wherein the profile circumferential evaluation index comprises the current pass relative reduction, the current pass relative spread, the circumferential unevenness coefficient of the profile circumference, the circumferential fluctuation coefficient of the profile circumference, the local weakness index of the profile circumference, the ear size, the current pass coating detection area proportion, the total area error rate, the core area error rate, the coating area error rate and the current pass elongation coefficient;

[0071] calculate the current pass relative reduction according to the profile height and the pre-rolling blank height;

[0072] According to the current pass relative width expansion amount, the relative width expansion amount of the current pass is calculated according to the cross-sectional area of the blank and the theoretical design cross-sectional area.

[0073] According to the maximum, minimum, mean and standard deviation of the cross-sectional length of the cross section of the blank, the circumferential unevenness coefficient of the cross section of the blank is calculated as the ratio of the maximum cross-sectional length to the minimum cross-sectional length, the circumferential fluctuation coefficient of the cross section of the blank is calculated as the ratio of the standard deviation of the cross-sectional length to the mean of the cross-sectional length, and the local weakness index of the cross section of the blank is calculated as the ratio of the minimum cross-sectional length to the mean of the cross-sectional length.

[0074] According to the cross-sectional area of the blank and the theoretical design cross-sectional area, the ear size of the cross section of the blank is calculated.

[0075] According to the total detection area and the total detection area, the proportion of the current pass coating detection area is calculated.

[0076] According to the total area error rate, the core area error rate and the coating area error rate are calculated according to the theoretical design total area and the total detection area, the theoretical design core area and the core detection area, and the theoretical design coating area and the coating detection area.

[0077] According to the cross-sectional area of the blank and the total detection area, the elongation coefficient of the current pass is calculated.

[0078] Further, the detection method further comprises calculating a blank coating evaluation index to further evaluate the characteristics of the blank coating; wherein the blank coating evaluation index comprises the circumferential unevenness coefficient of the coating, the circumferential fluctuation coefficient of the coating, the local weakness index of the coating, the number of peaks and the number of valleys.

[0079] According to the maximum, minimum, mean and standard deviation of the cross-sectional length of the cross section of the blank, the circumferential unevenness coefficient of the cross section of the blank is calculated as the ratio of the maximum cross-sectional length to the minimum cross-sectional length, the circumferential fluctuation coefficient of the cross section of the blank is calculated as the ratio of the standard deviation of the cross-sectional length to the mean of the cross-sectional length, and the local weakness index of the cross section of the blank is calculated as the ratio of the minimum cross-sectional length to the mean of the cross-sectional length.

[0080] According to the maximum, minimum, mean and standard deviation of the cross-sectional length of the cross section of the blank, the circumferential unevenness coefficient of the cross section of the blank is calculated as the ratio of the maximum cross-sectional length to the minimum cross-sectional length, the circumferential fluctuation coefficient of the cross section of the blank is calculated as the ratio of the standard deviation of the cross-sectional length to the mean of the cross-sectional length, and the local weakness index of the cross section of the blank is calculated as the ratio of the minimum cross-sectional length to the mean of the cross-sectional length.

[0081] Based on the same concept, the present application also provides a composite rod wire round hole blank cross-sectional feature intelligent detection system, the detection system comprises:

[0082] An acquisition unit is configured to acquire a calibration plate image and a cross-section image of a blank; and obtain an edge image from the cross-section image of the blank, wherein the edge image comprises an outer layer edge and a core edge;

[0083] A first calculation unit is configured to calculate a first image pixel scale from the calibration plate image;

[0084] A second calculation unit is configured to calculate a blank circumference length of each pixel point of the outer layer edge from the edge image;

[0085] A third calculation unit is configured to calculate a blank circumference width and its endpoint coordinates from the first image pixel scale and the leftmost and rightmost pixel points of the outer layer edge;

[0086] A fourth calculation unit is configured to calculate a blank height from the endpoint coordinates of the blank circumference width;

[0087] A fifth calculation unit is configured to calculate a cladding thickness of each pixel point of the core edge from the edge image and the blank circumference width;

[0088] A sixth calculation unit is configured to calculate a core detection area and a total detection area from the edge image, and calculate a cladding detection area from the core detection area and the total detection area.

[0089] Compared with the prior art, the present application has the following beneficial effects:

[0090] The present application can quantitatively describe the circumference length, width, height, cladding thickness and cladding area of the cross-section of the blank, and solve the problem of inaccurate detection of the cross-section characteristics of the round blank by manual field detection, and realize intelligent and high-precision detection of various characteristic data of the cross-section of the round blank of the composite rod wire, and provide accurate data support for the process and pass optimization of the composite rod wire.

[0091] The present application uses the actual width of the blank cross-section to verify and correct the detected blank circumference width, improves the detection accuracy of the blank width, solves the problem of inaccurate detection of the width and other characteristic data of the cross-section of the round blank of the composite rod wire during rolling, and realizes intelligent detection of the characteristic data of the cross-section of the round blank of the composite rod wire, with good stability, high precision and fast speed. BRIEF DESCRIPTION OF DRAWINGS

[0092] In order to more clearly illustrate the technical solutions of the present application, the following will briefly introduce the drawings needed in the embodiment description. Obviously, the drawings in the following description are only one embodiment of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0093] Figure 1 is a flow chart of the intelligent detection method of the cross-sectional characteristics of the composite bar wire round pass blank in embodiment one of the present application;

[0094] Figure 2 is a schematic diagram of the Φ12 mm composite spiral steel reinforcement hot rolling production line and blank in embodiment two of the present application;

[0095] Figure 3 is the image of the calibrated plate and the 1# flying shear sample F1 after corrosion in embodiment two of the present application;

[0096] Figure 4 is the image of the calibrated plate after pretreatment in embodiment two of the present application;

[0097] Figure 5 is the calibrated circle selected according to condition one and condition two in embodiment two of the present application;

[0098] Figure 6 is the core carbon steel filling image in embodiment two of the present application;

[0099] Figure 7 is the core carbon steel edge image in embodiment two of the present application;

[0100] Figure 8 is the outer layer edge image in embodiment two of the present application;

[0101] Figure 9 is the edge image in embodiment two of the present application;

[0102] Figure 10 is the feature data graph in embodiment two of the present application;

[0103] Figure 11 is the blank circumference length curve in embodiment two of the present application;

[0104] Figure 12 is the distance curve from the outer layer edge to the center point of the blank in embodiment two of the present application;

[0105] Figure 13 is the cladding thickness curve in embodiment two of the present application;

[0106] Figure 14 is the image of the calibrated plate and the Φ16 mm composite reinforcement 1# flying shear round pass blank S1 after corrosion in embodiment three of the present application;

[0107] Figure 15 is the cladding thickness change curve in embodiment three of the present application. DETAILED DESCRIPTION

[0108] The technical solutions in the present application will be described clearly and completely in the embodiments of the present application in conjunction with the drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all the other embodiments obtained by those skilled in the art without creative efforts belong to the scope of the present application.

[0109] The technical solutions of the present application will be described in detail in the specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes can not be described in some embodiments.

[0110] Embodiment one

[0111] As shown in the figure, the composite rod wire round hole type cross section feature intelligent detection method provided by the embodiment of the present application comprises the following steps: Figure 1

[0112] Step 1: Obtain the calibration plate image and the cross section image of the material type.

[0113] In the rolling process of the composite rod wire industrial site, the intermediate pass round hole material type is obtained by flying shear or emergency stop, and then the cross section sample of the material type is obtained along the normal plane of the rolling direction of the material type, the sample thickness is 5-10 mm, the sample is polished using a polishing machine, etched for 1-2 min using a 4% nitric acid alcohol etching solution, slowly washed in water and dried to obtain the sample.

[0114] Place the calibration plate on one side of the sample, and keep the width direction of the sample as horizontal as possible. Use a high-definition camera to capture the calibration plate and sample images. Use image processing software to extract the calibration plate image and the cross section image of the material type from the calibration plate and sample images.

[0115] The present application adopts the difference and uniform light source of bimetallic corrosion resistance, eliminates the influence of ambient light, and has strong anti-interference ability.

[0116] Step 2: Calculate the first image pixel ruler according to the calibration plate image.

[0117] In the specific embodiment of the present application, the first image pixel ruler is calculated according to the calibration plate image, which specifically comprises:

[0118] Step 2.1: Preprocess the calibration plate image; wherein the calibration plate image contains a plurality of calibration circles.

[0119] In this embodiment, the preprocessing of the calibration plate image comprises grayscale, median filtering and binary inversion processing in sequence.

[0120] ​Step 2.2: Extracting the multiple connected regions from the pre-processed calibration plate image, and measuring the area of each multiple connected region and the major axis length and minor axis length of the ellipse with the same normalized second central moment as each multiple connected region.

[0121] The multiple connected region in this embodiment is an 8-connected region, and the 8-connected region in the calibration plate image includes not only the calibration circle but also the bounding box and the like, and the number of 8-connected regions is greater than the number of calibration circles. In the MATLAB software, the function regionprops is used to measure the properties of each 8-connected region, such as the area, the centroid, the minimum rectangular frame, the major axis and the minor axis length of the region, and the like.

[0122] Step 2.3: Extracting the multiple connected region i which satisfies both condition one and condition two:

[0123] Condition one:

[0124] ;

[0125] Condition two:

[0126] ;

[0127] wherein, , respectively represent the area of the multiple connected region i and j; represents the area accuracy threshold; represents the number of multiple connected regions extracted from the pre-processed calibration plate image; and represents the number of multiple connected regions which satisfy both condition one and condition two. , respectively represent the major axis length and the minor axis length of the ellipse with the same normalized second central moment as the multiple connected region i.

[0128] In this embodiment, the area accuracy threshold is set to 0.1. Part of the calibration circles are screened out through the area limitation of condition one and the frame limitation of condition two.

[0129] Step 2.4: Calculating the first image pixel scale according to the extracted multiple connected regions which satisfy both condition one and condition two, and the specific calculation formula is as follows:

[0130] (1)

[0131] wherein, represents the first image pixel scale; represents the diameter of the calibration circle in the calibration plate; and n represents the number of the extracted multiple connected regions which satisfy both condition one and condition two.

[0132] Step 3: Obtaining the edge image according to the material type cross-sectional image.

[0133] ​​​​The core edge and the outer edge are extracted from the cross-sectional image of the material type by an edge detection algorithm, and the core edge and the outer edge are combined to obtain an edge image.

[0134] Step 4: The material type circumferential length of each pixel point of the outer edge is calculated according to the edge image.

[0135] In the specific embodiment of the present application, the material type circumferential length of each pixel point of the outer edge is calculated according to the edge image, and specifically includes:

[0136] Step 4.1: The leftmost horizontal pixel point of the outer edge is taken as the first point and recorded as Na'(1), all pixel points of the outer edge are sorted, and the serial number of the highest vertical pixel point is recorded as Nb', the serial number of the rightmost horizontal pixel point is recorded as Nc', and the serial number of the lowest vertical pixel point is recorded as Nd'.

[0137] Step 4.2: The distance between each pixel point of the outer edge and other pixel points is calculated, and the specific calculation formula is:

[0138] (2)

[0139] wherein, represents the distance between the i-th pixel point and the j-th pixel point of the outer edge, , , represents the number of pixel points of the outer edge; represents the coordinates of the i-th pixel point of the outer edge; represents the coordinates of the j-th pixel point of the outer edge.

[0140] Step 4.3: The maximum distance between each pixel point of the outer edge and other pixel points is found out, and it is judged whether the maximum distance is greater than the first reference distance, if not, the first reference distance is taken as the maximum distance.

[0141] For the i-th pixel point of the outer edge, the maximum value in is found out and recorded as , , which represents the maximum distance between the i-th pixel point of the outer edge and other pixel points. If is less than the first reference distance, the first reference distance is taken as .

[0142] In this embodiment, the determination method of the first reference distance is:

[0143] If all the pixel points of the outer edge are sorted in a clockwise direction, and i

[0144] If all the pixel points of the outer edge are sorted in an anticlockwise direction, and i

[0145] Step 4.4: the maximum distance between each pixel point of the outer edge and other pixel points The length of the material type circle for each pixel point of the outer edge.

[0146] The length of the material type circle for each pixel point of the outer edge. The maximum value is found from the length of the material type circle of all the pixel points of the outer edge and is recorded as , , which represents the maximum length of the material type circle, and the pixel point sequence number corresponding to the maximum length of the material type circle is determined , Similarly, the minimum length of the material type circle and the pixel point sequence number corresponding thereto are found.

[0147] Step 5: calculating the material type circle width and its endpoint coordinates according to the first image pixel scale and the leftmost horizontal pixel point and the rightmost horizontal pixel point of the outer edge.

[0148] In the specific embodiments of the present application, calculating the material type circle width and its endpoint coordinates according to the first image pixel scale and the leftmost horizontal pixel point and the rightmost horizontal pixel point of the outer edge comprises:

[0149] Step 5.1: taking the leftmost horizontal pixel point of the outer edge as the first point and recording it as Na'(1), sorting all the pixel points of the outer edge, and recording the sequence number of the uppermost vertical pixel point as Nb', the sequence number of the rightmost horizontal pixel point as Nc', and the sequence number of the lowermost vertical pixel point as Nd'.

[0150] Step 5.2: determining the line between the leftmost horizontal pixel point and the rightmost horizontal pixel point of the outer edge, i.e. connecting the point labeled as Na'(1) and the point labeled as Nc'.​

[0151] Step 5.3: According to the sequence number deviation between the end point sequence number of the circumferential spread of the material shape and the horizontal leftmost pixel point sequence number Na'(1) and the horizontal rightmost pixel point sequence number Nc' of the outer edge, the end point sequence number of the circumferential spread of the material shape is determined.

[0152] During the composite rod wire rolling process, the material shape of the round pass is a flat ellipse or a vertical ellipse, and the line between the horizontal leftmost pixel point and the horizontal rightmost pixel point of the outer edge is basically coincided with the circumferential spread or the height of the material shape, but there is a certain error. Taking the 1# flying shear machine as an example, the material shape is a flat ellipse, and the line between the horizontal leftmost pixel point and the horizontal rightmost pixel point of the outer edge is basically consistent with the circumferential spread of the material shape. The horizontal leftmost pixel point sequence number Na'(1) and the horizontal rightmost pixel point sequence number Nc' of the outer edge and the line are drawn in the material shape cross-sectional image, and the sequence number deviation between the end point sequence number of the circumferential spread of the material shape and the horizontal leftmost pixel point sequence number Na'(1) and the horizontal rightmost pixel point sequence number Nc' of the outer edge is determined through observation and analysis. 、 The end point sequence number of the circumferential spread of the material shape is:

[0153] , (3)

[0154] Wherein, 、 respectively represent the left end point sequence number and the right end point sequence number of the circumferential spread of the material shape. 、 When the value is positive, it means adjusting clockwise and upward; 、 When the value is negative, it means adjusting counterclockwise and downward.

[0155] Step 5.4: According to the end point coordinates of the circumferential spread of the material shape and the first image pixel scale, the circumferential spread of the material shape is calculated, and the specific calculation formula is:

[0156] (4)

[0157] Wherein, represents the circumferential spread of the material shape; 、 respectively represent the coordinates of the two end points of the circumferential spread of the material shape; represents the first image pixel scale.

[0158] In order to improve the detection accuracy of the circumferential spread of the material shape and other characteristic data, the circumferential spread of the material shape is also corrected, which specifically includes:

[0159] Step 5.5: Obtain the actual width of the cross section of the material, and calculate the width error rate according to the actual width of the cross section of the material and the circumferential width of the material.

[0160] The actual width of the cross section of the sample in step 1 is measured by using a vernier caliper, and the calculation formula of the width error rate is as follows:

[0161] (5)

[0162] Wherein, represents the width error rate, and B represents the actual width of the cross section of the material.

[0163] Step 5.6: Determine whether the width error rate is less than the error rate threshold value, if yes, the circumferential width of the material calculated in step 5.5 has high reliability and accuracy, and does not need to be corrected; if not, calculate the second image pixel ruler according to the end point coordinates of the circumferential width of the material and the actual width of the cross section of the material, and the specific calculation formula is as follows: (6)

[0164] (7)

[0165] Wherein, represents the second image pixel ruler.

[0166] Step 5.7: Calculate the corrected circumferential width of the material according to the second image pixel ruler and the end point coordinates of the circumferential width of the material, and the specific formula is as follows:

[0167] (7)

[0168] Wherein, represents the corrected circumferential width of the material. The corrected circumferential width of the material is basically equal to the actual width of the cross section of the material B.

[0169] Step 6: Calculate the height of the material according to the end point coordinates of the circumferential width of the material.

[0170] In the specific embodiment of the present application, the height of the material is calculated according to the end point coordinates of the circumferential width of the material, and specifically includes:

[0171] Step 6.1: Determine the horizontal coordinate of the midpoint of the width direction according to the end point coordinates of the circumferential width of the material.

[0172] Take the left end point of the circumferential width of the material as the first point, sort all the pixel points of the outer edge, and the sequence number of the right end point of the circumferential width of the material is recorded as n. Then the horizontal coordinate of the midpoint of the width direction is:

[0173] ​​​(8)

[0174] wherein, represents the abscissa of the midpoint of the spread direction; represents the abscissa of the left end point of the circumferential spread of the material type; represents the abscissa of the right end point of the circumferential spread of the material type; represents a rounding function.

[0175] Step 6.2: determining the abscissa value range according to the set abscissa deviation and the abscissa of the midpoint.

[0176] The set abscissa deviation is denoted as , and the abscissa value range is ( , ).

[0177] Step 6.3: determining the upper intersection point sequence number and the lower intersection point sequence number of the vertical line passing through each abscissa in the abscissa value range and the upper edge, and further determining the corresponding ordinate of each abscissa in the abscissa value range and the midpoint of the spread direction.

[0178] Each abscissa in the abscissa value range is denoted as , , , the vertical line passing through each abscissa intersects the upper edge, and the upper intersection point sequence number and the lower intersection point sequence number are denoted as , , respectively. The corresponding ordinate of each abscissa is , , wherein represents the ordinate of the upper intersection point and the lower intersection point of the vertical line passing through the abscissa .

[0179] Step 6.4: determining the upper groove bottom arc center point coordinate corresponding to each midpoint according to the vertical distance difference between the upper groove bottom arc center point of the roll pass and the actual midpoint of the spread direction, and determining the lower groove bottom arc center point coordinate corresponding to each midpoint according to the vertical distance difference between the lower groove bottom arc center point of the roll pass and the actual midpoint of the spread direction.

[0180] The vertical distance difference between the upper groove bottom arc center point of the roll pass and the actual midpoint of the spread direction is equal to the vertical distance difference between the lower groove bottom arc center point of the roll pass and the actual midpoint of the spread direction, which is denoted as , and the vertical distance difference can be obtained according to the roll pass diagram.

[0181] ​The coordinates of the upper groove bottom arc center point O1(k) corresponding to each intermediate point k are The coordinates of the lower groove bottom arc center point O2(k) corresponding to each intermediate point k are wherein PX represents an image pixel scale, if the circumferential width of the material type meets the accuracy requirement, no correction is needed, and PX is ; if the circumferential width of the material type does not meet the accuracy requirement, correction is needed, and PX is .

[0182] Step 6.5: Set the value range of the upper intersection point sequence number and the lower intersection point sequence number.

[0183] In this embodiment, it is assumed that the value range of the upper intersection point sequence number and the lower intersection point sequence number of the vertical line passing through each abscissa value in the abscissa value range and the outer edge is K, the upper intersection point sequence number is , and the lower intersection point sequence number is . .

[0184] Step 6.6: Calculate the distance between each upper intersection point in the value range of the upper intersection point sequence number and the corresponding upper groove bottom arc center point, and the distance between each lower intersection point in the value range of the lower intersection point sequence number and the corresponding lower groove bottom arc center point, to form a distance set composed of all distances.

[0185] Step 6.7: Calculate the variance of each distance in the distance set.

[0186] Step 6.8: Select the intermediate point corresponding to the distance with the smallest variance as the material type center point, that is, select k corresponding to the distance with the smallest variance, and the k at this time is recorded as . The corresponding abscissa and ordinate are taken as the material type center point, that is, the coordinates of the material type center point are . .

[0187] Step 6.9: Determine the upper intersection point and the lower intersection point of the vertical line passing through the material type center point and the outer edge, and calculate the material type height according to the upper intersection point and the lower intersection point, and the specific calculation formula is:

[0188] (9)

[0189] wherein represents the material type height; and respectively represent the upper intersection point and the lower intersection point of the vertical line passing through the material type center point and the outer edge. The pixel scale selected in step 5 when calculating the circumferential width of the material type is taken as the pixel scale when calculating the material type height. The height error rate can be calculated according to the material type height and the actual height measured.

[0190] In this embodiment, the distance from the center point of the material to each pixel point of the outer edge is also calculated. The specific calculation formula is:

[0191] (10)

[0192] in, Indicates the distance from the center point of the material to each pixel point of the outer edge; Pixels representing the outer edge; Indicates the center point of the material. PX in formula (10) is determined by PX in formula (9), that is, if the material height is based on Calculated, then PX of formula (10) is ; If the material height is based on Calculated, then PX of formula (10) is .

[0193] Step 7: Calculate the coating thickness of each pixel at the edge of the core based on the edge image and the circumferential width of the material.

[0194] In a specific embodiment of the present invention, the cladding thickness of each pixel at the edge of the core is calculated based on the edge image and the circumferential width of the material, including:

[0195] Step 7.1: Take the intersection point of the straight line where the circumference of the material is widened and the left side of the core edge and the outer edge as the first point of the core edge and the outer edge, and sort all the pixel points of the core edge and the outer edge respectively.

[0196] For example, the intersection point of the straight line where the material mold circumference is widened and the left side of the core edge is taken as the first point of the core edge, and all pixel points on the core edge are sorted in a clockwise direction; the intersection point of the straight line where the material mold circumference is widened and the left side of the outer edge is taken as the first point of the outer edge, and all pixel points on the outer edge are sorted in a clockwise direction.

[0197] Step 7.2: Take the distance between the first point on the edge of the core and the first point on the edge of the outer layer as the second reference distance.

[0198] Step 7.3: Calculate the distance between each pixel point of the core edge and other pixel points of the outer edge; wherein, other pixel points of the outer edge refer to pixel points other than the first pixel point of the outer edge.

[0199] Step 7.4: Find the minimum distance between each pixel on the edge of the core and any other pixel on the edge of the outer layer.

[0200] For each pixel point of the core edge, find the minimum value from the distances between the pixel point and each other pixel point of the outer edge, and the minimum value is the minimum distance between the pixel point and the other pixel point of the outer edge.

[0201] Step 7.5: Determine whether the minimum distance between each pixel point of the core edge and the other pixel point of the outer edge is less than the second reference distance.

[0202] If the minimum distance is less than the second reference distance, the minimum distance is the coating thickness of the corresponding pixel point of the core edge; if the minimum distance is greater than or equal to the second reference distance, the second reference distance is the coating thickness of the corresponding pixel point of the core edge. Thus, the coating thickness of each pixel point of the core edge can be calculated.

[0203] Step 8: Calculate the core detection area and the total detection area according to the edge image, and calculate the coating detection area according to the core detection area and the total detection area.

[0204] Fill the core edge in the edge image, and then calculate the filled area, which is the core detection area; fill the outer edge in the edge image, and then calculate the filled area, which is the total detection area; the coating detection area is equal to the difference between the total detection area and the core detection area.

[0205] Step 9: Calculate the material type circumference evaluation index, and then evaluate the material type circumference characteristics.

[0206] In this embodiment, the material type circumference evaluation index includes the current pass relative reduction, the current pass relative spread, the hoop unevenness coefficient of the material type circumference, the hoop fluctuation coefficient of the material type circumference, the local weakness index of the material type circumference, the ear size, the current pass coating detection area proportion, the total area error rate, the core area error rate, the coating area error rate, and the current pass elongation coefficient.

[0207] The current pass relative reduction is calculated according to the material type height and the pre-rolling blank height, and the specific calculation formula is:

[0208] (11)

[0209] Wherein, represents the current pass relative reduction; represents the pre-rolling blank height, which is measured; h represents the material type height, which is calculated in step 6.

[0210] The current pass relative spread is calculated according to the material type circumference spread and the pre-rolling blank spread, and the specific calculation formula is:

[0211] (12)

[0212] in, Indicates the relative width of the current pass; Indicates the circumferential width of the material, which is calculated in step 5. When the circumferential width of the material needs to be corrected, the circumferential width of the material here is the circumferential width of the material after correction; Indicates the width of the billet before rolling, obtained by measurement.

[0213] According to the circumference length of the material mold at each pixel point on the outer edge, the maximum circumference length of the material mold, the minimum circumference length of the material mold, the mean circumference length of the material mold and the standard deviation of the circumference length of the material mold are calculated, and the ratio of the maximum circumference length of the material mold to the minimum circumference length of the material mold is used as the circumferential unevenness coefficient of the material mold circumference, the ratio of the standard deviation of the material mold circumference length to the mean circumference length of the material mold is used as the circumferential fluctuation coefficient of the material mold circumference, and the ratio of the minimum circumference length of the material mold to the mean circumference length of the material mold is used as the local weakness index of the material mold circumference.

[0214] In this embodiment, when the circumferential unevenness coefficient of the material mold circumference is less than 1.3, it indicates that the material mold circumferentially is uniform; when the circumferential unevenness coefficient of the material mold circumference is greater than or equal to 1.3, it indicates that the material mold circumferentially is uneven. When the circumferential fluctuation coefficient of the material mold circumference is less than 0.1, it indicates that the circumferential fluctuation of the material mold circumference is small; when the circumferential unevenness coefficient of the material mold circumference is greater than or equal to 0.1, it indicates that the circumferential fluctuation of the material mold circumference is large. When the local weakness index of the material mold circumference is greater than 0.8, it indicates that the material mold circumference is locally not weak; when the local weakness index of the material mold circumference is less than or equal to 0.8, it indicates that the material mold circumference is locally weak.

[0215] The ear size of the cross section of the material is calculated based on the circumferential width of the material and the theoretical design width. The specific calculation formula is:

[0216] (13)

[0217] in, Indicates the ear size of the material cross section; Indicates the circumferential width of the material, which is calculated in step 5. When the circumferential width of the material needs to be corrected, the circumferential width of the material here is the circumferential width of the material after correction; Indicates theoretical design expansion.

[0218] The proportion of the current pass's coating inspection area is calculated based on the coating inspection area and the total inspection area. The specific calculation formula is:

[0219] (14)

[0220] in, Indicates the current pass coating detection area ratio, Indicates the coating detection area, Indicates the total detection area.

[0221] Let the outer diameter of the outer layer of the composite rod be , the thickness be , the cross-sectional area of the outer layer of the initial blank be , and the cross-sectional area of the initial blank be , then the calculation formula of the cross-sectional area ratio of the outer layer is:

[0222] (15)

[0223] wherein, represents the cross-sectional area ratio of the outer layer.

[0224] According to the total area of the theoretical design and the total detection area, the total area error rate Serr is calculated, according to the core area of the theoretical design and the core detection area, the core area error rate Sterr is calculated, and according to the coating area of the theoretical design and the coating detection area, the coating area error rate Sberr is calculated.

[0225] According to the cross-sectional area of the blank and the total detection area, the current pass elongation coefficient is calculated, and the specific calculation formula is:

[0226] (16)

[0227] wherein, represents the current pass elongation coefficient.

[0228] Step 10: Calculate the material type coating evaluation index, and then evaluate the material type coating characteristics.

[0229] In this embodiment, the material type coating evaluation index includes the hoop unevenness coefficient of the coating, the hoop fluctuation coefficient of the coating, the local weakness index of the coating, the number of wave crests, and the number of wave troughs.

[0230] According to the coating thickness of each pixel point of the core edge, the maximum coating thickness, the minimum coating thickness, the coating thickness mean value, and the coating thickness standard deviation are calculated, the ratio of the maximum coating thickness to the minimum coating thickness is taken as the hoop unevenness coefficient of the coating, the ratio of the coating thickness standard deviation to the coating thickness mean value is taken as the hoop fluctuation coefficient of the coating, and the ratio of the minimum coating thickness to the coating thickness mean value is taken as the local weakness index of the coating.

[0231] In this embodiment, when the circumferential unevenness coefficient of the coating is less than 1.3, it indicates that the coating is circumferentially uniform; when the circumferential unevenness coefficient of the coating is greater than or equal to 1.3, it indicates that the coating is circumferentially uneven. When the circumferential fluctuation coefficient of the coating is less than 0.1, it indicates that the circumferential fluctuation of the coating is small; when the circumferential unevenness coefficient of the coating is greater than or equal to 0.1, it indicates that the circumferential fluctuation of the coating is large. When the local weakness index of the coating is greater than 0.8, it indicates that the coating is locally not weak; when the local weakness index of the coating is less than or equal to 0.8, it indicates that the coating is locally weak.

[0232] To better characterize the uniformity of the cladding thickness of the material, a cladding thickness curve is plotted based on the cladding thickness at each pixel point on the core edge. A horizontal line (average ± deviation of the cladding thickness) is drawn within the cladding thickness curve. The number of peaks and troughs is determined based on the horizontal line. The cladding uniformity is evaluated based on the number of peaks and troughs.

[0233] The intelligent detection system for cross-sectional characteristics of round hole-shaped composite rod and wire provided by the embodiment of the present invention includes:

[0234] An acquisition unit is used to acquire a calibration plate image and a material cross-sectional image; obtain an edge image according to the material cross-sectional image, wherein the edge image includes an outer layer edge and a core edge;

[0235] A first calculation unit, configured to calculate a first image pixel scale according to the calibration plate image;

[0236] A second calculation unit is used to calculate the circumference length of the material mold at each pixel point of the outer edge according to the edge image;

[0237] A third calculation unit is used to calculate the circumferential width of the material mold and the coordinates of its endpoints based on the pixel scale of the first image and the horizontal leftmost pixel point and the horizontal rightmost pixel point of the outer edge;

[0238] A fourth calculation unit is used to calculate the height of the material mold according to the coordinates of the endpoints of the circumferential width of the material mold;

[0239] A fifth calculation unit is used to calculate the coating thickness of each pixel point at the edge of the core according to the edge image and the circumferential width of the material;

[0240] The sixth calculation unit is configured to calculate a core detection area and a total detection area according to the edge image, and calculate a cladding detection area according to the core detection area and the total detection area.

[0241] In some specific embodiments of the present invention, the intelligent detection system for cross-sectional features of composite rod and wire round hole material can be combined with the features of the intelligent detection method for cross-sectional features of composite rod and wire round hole material in Example 1 of the present invention, and vice versa.

[0242] Example 2

[0243] Coiled rebar refers to a coiled threaded steel bar with a diameter of 6-12 mm rolled on a high-speed wire production line. Among the steel bars for concrete structures, about 20-25% are fine-diameter steel bars with a nominal diameter of less than 12 mm, which are mainly used as load-bearing steel bars for components such as slabs and walls, and as stirrups, ties, and structural bars in beam and column components.

[0244] Stainless steel composite coiled rebar, hereinafter referred to as composite coiled rebar, is a composite bar wire with an outer layer of stainless steel and a core of carbon steel. It has high strength, high toughness, and excellent corrosion resistance, and can also save valuable nickel and chromium resources in stainless steel, thereby improving economic benefits and having broad development prospects in engineering applications.

[0245] First, standard 316L stainless steel seamless pipes (Φ159x9000, wall thickness 6mm, conforming to GB / T20878-2007) and HRB400E carbon steel core rods (Φ155x9000mm, machined to Φ146.90mm) are purchased to prepare composite round billets.

[0246] A Φ12mm composite coiled rebar hot rolling industrial test is carried out in the high-speed wire mill of a certain steel plant. The production line has a total of 22 rolling mills (1-22#), which are horizontally and vertically distributed in sequence. The rough, medium, pre-precision rolling, and precision rolling have 6-6-6-4 passes, respectively. The flying shear machines of the rough, medium, and pre-precision rolling are between 6-7#, 12-13#, and 18-19# rolling mills, respectively, and are sequentially numbered as 1-3# flying shears, as shown in Figure 2 . Figure 2 Among them, 1 represents the calibration circle, 2 represents the calibration circle with a diameter of 4mm, 3 represents the core edge, 4 represents the outer layer edge, and 5 represents the lug. Box-type holes are used for 1-2# rolling mills, elliptical-circular holes are used for 3-22# rolling mills, and the finished product hole of 22# rolling mill meets the requirements of GB / T1499.2-2018. The Φ12mm composite coiled rebar product is rolled, and the industrial field hot rolling process is as follows:

[0247] Heating temperature and time: 1060±40℃ in the soaking section, heating for 2.5h; rolling temperature: greater than 1025±30℃; sample acquisition: 1-3# flying shear samples (F1, F2, and F3) are obtained on site by emergency stop, and Φ12mm composite coiled rebar finished product Fp is obtained on the cooling bed; sample cooling method: 1-3# flying shear samples are air-cooled to room temperature, and the finished product is water-cooled to room temperature.

[0248] Taking the 1# flying shear sample F1 in the hot rolling process of Φ12mm composite coiled rebar as an example, the following is explained and described:

[0249] 1# flying shear sample F1 was etched with 4% nitric acid for 1-2 minutes, then slowly rinsed in water and dried. The calibration circle diameter D0 was 4 mm, and a yellow PVC board was used as the background plate for the sample cell.

[0250] like Figure 3 The images shown are the calibration plate and the 1# flying shear sample F1 after corrosion. In addition, the actual width B of the cross section measured with a vernier caliper is 83.64mm, and the actual height H is 75.54mm. Figure 3 The image is cropped and segmented into the calibration plate image and the material cross-sectional image. Figure 4 The pre-processed calibration plate image shows 49 calibration circles and the area accuracy threshold. Set to 0.1, and select 18 calibration circles through the area limitation of condition 1 and the frame limitation of condition 2 in embodiment 1, such as Figure 5 According to step 2 in embodiment 1, the first image pixel scale is calculated. It is 0.0414mm / pixel.

[0251] The cross-sectional image of the corroded material is first binarized, closed, and filled to obtain the core carbon steel filling image, such as Figure 6 As shown, edge detection is then performed to obtain the core carbon steel edge image, as shown Figure 7 shown.

[0252] The cross-sectional image of the corroded material is contrast-enhanced and opened to obtain the background image. The difference between the two is used to obtain the foreground image and binarized to remove Figure 6 The core carbon steel filling area is shown, and then the connected domain with the largest area is found. This connected domain is the stainless steel cladding. After filling the cladding, its outer edge is obtained, as shown in Figure 8 The core carbon steel edge image and the stainless steel cladding outer edge image are combined, and the result is shown in Figure 9 As shown, the edge image is obtained.

[0253] The circumferential width b of the material calculated according to step 5 of Example 1 is 86.13 mm, and the width error rate is The circumferential width b of the material is corrected, and the pixel scale of the second image is calculated at this time. The left end point of the circumference of the corrected material is taken as the first point, and all the pixel points on the outer edge are reordered and the pixel scale of the second image is used. , the height of the material shape, the center point of the material shape, the distance from each pixel point of the outer edge to the center point of the material shape are calculated according to step 6 in the embodiment one of the present application, and then the circumferential length of each pixel point of the outer edge is recalculated according to step 4 in the embodiment one, the height of the material shape, the center point of the material shape, the distance from each pixel point of the outer edge to the center point of the material shape are calculated according to step 6 in the embodiment one, the thickness of the coating of each pixel point of the core edge is calculated according to step 7 in the embodiment one, and so on, and the drawing is plotted according to these calculated characteristic data, as shown in Figure 10 . Figure 10 In the drawing, 6 represents the clockwise direction, 7 represents the line connecting the leftmost horizontal pixel point and the rightmost horizontal pixel point of the outer edge, 8 represents the circumferential spread line of the material shape, 9 represents the minimum circumferential length of the material shape, 10 represents the height of the material shape, 11 represents the maximum thickness of the coating, 12 represents the minimum thickness of the coating, 13 represents the upper groove bottom arc, 14 represents the lower groove bottom arc, and 15 represents the horizontal line connecting the ends of the tangent lines on both sides of the upper groove bottom (the tangent lines connecting the groove bottom arcs in the round hole type). Table 1 shows the obtained characteristic data.

[0254] Table 1 Φ12 composite disc screw 1# flying shear round hole material cross-sectional characteristic data

[0255]

[0256] The corrected circumferential spread b' of the material shape in the present embodiment is 83.64 mm, which is equal to the actual spread B of the cross section; the height h of the material shape calculated through the second image pixel ruler is 75.32 mm, and the error rate with respect to the actual height H is 0.29%, which meets the accuracy requirement.

[0257] The core detection area and the total detection area are 3872.69 mm 2 and 4570.70 mm 2 respectively, so the coating detection area is equal to 698.01 mm 2 . The height of the blank before rolling is 159 mm, and the relative reduction in the current pass is 52.63%; the spread of the blank before rolling is 159 mm, and the relative spread in the current pass is -47.38%; the designed spread is 75.3, and the ear size is 4.17 mm.

[0258] The proportion of the outer cross-sectional area is 14.52%, and the proportion of the coating detection area in the current pass is 15.27%, which can be used for coating rheological difference analysis.

[0259] According to the area calculation method of the round hole material shape in the rolling theory, the total designed area of the F1 sample of the current 1# flying shear is 4547.00 mm 2According to the theoretical design of proportional calculation, the core area is 3881.28mm 2 , the theoretical design covering area is 660.44mm 2 According to the theoretical design of coating area and coating detection area, the coating detection area of ​​F1 sample of 1# flying shear increases by 37.57mm. 2 The total area error rate, core area error rate and cladding area error rate of the current pass are calculated to be 0.52%, 0.22% and 5.7% respectively. The total inspection area is 4570.70mm 2 , the current pass elongation coefficient of the rolling process is 4.3441.

[0260] Draw the circumference of the material and the distance from the outer edge to the center of the material And the coating thickness variation curve, such as Figure 11 to Figure 13 As shown, Nb', Nc', and Nd' respectively represent the vertical highest pixel point, the horizontal rightmost pixel point, and the vertical lowest pixel point of the outer edge, and Nb' 芯 、Nc' 芯 、Nd' 芯 They represent the vertical highest pixel point, the horizontal rightmost pixel point, and the vertical lowest pixel point of the core edge, respectively. FC The quantitative evaluation indicators of the circumferential size of the material and the coating unevenness are shown in Table 2, which can provide data support for the subsequent composite rod and wire rolling.

[0261] Table 2 Quantitative evaluation indicators of material circumference size and coating unevenness

[0262]

[0263] Example 3

[0264] Standardized 304 stainless steel welded pipes with an outer diameter of Φ168×9000mm and a wall thickness of 8mm were purchased from the market, and the welds were welded by 304 stainless steel electrodes; HRB400E carbon steel core rods with a size of Φ155×9000mm were processed to Φ151.90mm to prepare composite round billets.

[0265] A small-batch production trial of Φ16mm composite steel bars was conducted at a steel mill. The mills comprised 18 rolling mills (designated 1–18#), arranged horizontally and vertically. Roughing, intermediate, and finishing rolling were performed in six passes each. The roughing and intermediate rolling shears, designated 1# and 2#, were located between mills 6–7# and 12–13#, respectively. During the rolling process, the head and tail of the composite round billet were removed. Shear specimens 1# and 2#, designated S1 and S2, were obtained at the rolling site. Finished specimens Sp were also obtained on the cooling bed. Cross-sectional and longitudinal sections of the finished composite steel bar specimens Sp revealed significant unevenness in the coating.

[0266] The 1# flying shear sample S1 is corroded by 4% nitric acid alcohol solution for 1-2 min, slowly washed in water and dried. The calibration circle diameter D0 can be selected as 4 mm, and the sample pool selects a yellow PVC plate as the background plate, as shown in the collected calibration plate and 1# flying shear round hole type material S1 after corrosion. Figure 14 The actual cross-sectional width B is measured by a vernier caliper as 82.60 mm, and the actual height H is 76.48 mm.

[0267] The corrected material width b' is 82.60 mm, the corrected material height h' is 75.82 mm, and the height error rate is 0.86%, so the height measurement error after correction is very small. The maximum value, minimum value, mean value and standard deviation of the coating thickness are 5.24 mm, 2.10 mm, 3.61 mm and 0.5930 mm respectively, and the coating thickness change curve is shown in Figure 15 , wherein D FC represents the coating thickness. The S1 material coating evaluation indexes are: the ring unevenness coefficient of the coating is 2.495, the ring fluctuation coefficient of the coating is 0.164, the local weakness index of the coating is 0.582, and the number of wave peaks and wave troughs is 15.

[0268] The core detection area and the total detection area are 3784.78 mm 2 and 4635.66 mm 2 , respectively, so the coating detection area is equal to 850.88 mm 2 . The blank height before rolling is 168 mm, and the relative reduction of the current pass is 54.87%; the blank width before rolling is 168 mm, and the relative width reduction of the current pass is -50.83%.

[0269] The outer layer cross-sectional area ratio is 18.14%, and the current pass coating detection area ratio is 18.36%, which can be used for coating rheological difference analysis. The total detection area is 4635.66 mm 2 , and the elongation coefficient of the current pass in the rolling process is 4.7818. The elongation coefficients of the 1-6# rolling mills in the industrial hot rolling field are: 1.23x1.220x1.365x1.292x1.288x1.339=4.5641. The actual elongation is greater than the field test value, which provides data support for the optimization of the subsequent industrial production elongation coefficient.

[0270] The above only discloses specific embodiments of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or modifications within the technical scope disclosed by the present application, which should be covered within the protection scope of the present application.

Claims

1. An intelligent detection method for cross-sectional characteristics of round hole-shaped composite rod and wire material, characterized in that: The detection method comprises: Acquire calibration plate image and material cross-sectional image; Calculating a first image pixel scale according to the calibration plate image; Obtaining an edge image according to the cross-sectional image of the material, wherein the edge image includes an outer layer edge and a core edge; Calculate the circumference length of the mold at each pixel point on the outer edge according to the edge image; Calculate the circumferential width of the mold and its endpoint coordinates based on the pixel scale of the first image and the horizontal leftmost pixel point and the horizontal rightmost pixel point of the outer edge; Calculate the material height according to the coordinates of the endpoints of the material circumference; Calculate the coating thickness of each pixel point on the core edge based on the edge image and the circumferential width of the material; A core detection area and a total detection area are calculated according to the edge image, and a cladding detection area is calculated according to the core detection area and the total detection area.

2. The intelligent detection method for cross-sectional characteristics of round hole profile of composite rod and wire according to claim 1 is characterized in that: The acquisition process of the calibration plate image and the material cross-sectional image is as follows: Obtain cross-sectional specimens of the material; The cross-section specimen of the material is polished using a polishing machine, and then the polished cross-section specimen is corroded using a 4% nitric acid alcohol corrosive solution, and then cleaned and dried to obtain a sample; Place the calibration plate on one side of the sample, keeping the width of the sample horizontal, and capture images of the calibration plate and sample; The calibration plate and sample images are cropped and segmented to obtain the calibration plate image and the material cross-sectional image.

3. The intelligent detection method for cross-sectional characteristics of round hole profile of composite rod and wire according to claim 1, characterized in that: Calculating the pixel scale of the first image according to the calibration plate image includes: Preprocessing the calibration plate image; wherein the calibration plate image includes a plurality of calibration circles; Extract multi-connected regions from the pre-processed calibration plate image, and measure the area of ​​each multi-connected region and the major axis length and minor axis length of the ellipse with the same normalized second-order central moment as each multi-connected region; Extract the multiply connected region i that satisfies both conditions 1 and 2: Condition 1: ; Condition 2: ; in, 、 Represent the areas of multiply connected regions i and j respectively; Indicates the area accuracy threshold; represents the number of multi-connected regions extracted from the preprocessed calibration plate image; 、 They represent the major axis length and minor axis length of the ellipse with the same normalized second-order central moment as the multiply connected region i; The pixel scale of the first image is calculated based on the extracted multi-connected regions that meet both conditions 1 and 2. The specific calculation formula is: ； in, represents the first image pixel scale; Represents the diameter of the calibration circle in the calibration plate; n represents the number of multiply connected regions that meet both conditions 1 and 2.

4. The intelligent detection method for cross-sectional characteristics of round hole profile of composite rod and wire according to claim 1, characterized in that: Calculating the circumference length of the material mold at each pixel point on the outer edge according to the edge image includes: Take the leftmost horizontal pixel of the outer edge as the first point, sort all the pixels of the outer edge, and record the vertical highest pixel number as Nb', the horizontal rightmost pixel number as Nc', and the vertical lowest pixel number as Nd'; Calculate the distance between each pixel point on the outer edge and other pixel points. The specific calculation formula is: ； in, Represents the distance between the i-th pixel and the j-th pixel on the outer edge, , , Indicates the number of pixels at the outer edge; Represents the coordinates of the i-th pixel point on the outer edge; Represents the coordinates of the j-th pixel point on the outer edge; Find the maximum distance between each pixel point on the outer edge and other pixel points, and determine whether the maximum distance is greater than a first reference distance. If not, use the first reference distance as the maximum distance; wherein the first reference distance is determined as follows: If all pixels on the outer edge are sorted in a clockwise direction and i < Nc', then the first reference distance is the distance between the i-th pixel and the lowest vertical pixel; if all pixels on the outer edge are sorted in a clockwise direction and i > Nc', then the first reference distance is the distance between the i-th pixel and the highest vertical pixel; If all pixels on the outer edge are sorted in a counterclockwise direction and i < Nc', then the first reference distance is the distance between the i-th pixel and the highest vertical pixel; if all pixels on the outer edge are sorted in a counterclockwise direction and i > Nc', then the first reference distance is the distance between the i-th pixel and the lowest vertical pixel; The maximum distance between each pixel point on the outer edge and other pixel points is the circumference length of the material mold of each pixel point on the outer edge.

5. The intelligent detection method for cross-sectional characteristics of round hole profile of composite rod and wire according to claim 1, characterized in that: Calculating the circumferential width of the mold and its endpoint coordinates according to the first image pixel scale and the horizontal leftmost pixel point and the horizontal rightmost pixel point of the outer edge, including: taking the horizontal leftmost pixel point of the outer edge as the first point, sorting all the pixel points of the outer edge, and recording the sequence number of the horizontal rightmost pixel point as Nc'; Determine a line connecting the horizontal leftmost pixel point and the horizontal rightmost pixel point of the outer edge; determine the deviation between the endpoint number of the circumferential width of the material mold and the sequence number of the horizontal leftmost pixel point and the sequence number of the horizontal rightmost pixel point of the outer edge based on the line, and then determine the endpoint number of the circumferential width of the material mold; The circumferential width of the material is calculated based on the endpoint coordinates of the circumferential width of the material and the pixel scale of the first image. The specific calculation formula is: ； in, Indicates the circumference of the material is wide; 、 Respectively represent the coordinates of the two end points of the circumference of the material; Indicates the pixel scale of the first image.

6. The intelligent detection method for cross-sectional characteristics of round hole profile of composite rod and wire according to claim 5, characterized in that: The detection method further includes correcting the circumferential width of the material, specifically including: Obtain the actual width of the material cross section, and calculate the width error rate based on the actual width of the material cross section and the circumferential width of the material; Determine whether the widening error rate is less than the error rate threshold. If not, calculate the second image pixel scale based on the endpoint coordinates of the circumferential widening of the material and the actual widening of the cross section of the material. The specific calculation formula is: ； in, represents the second image pixel scale; Indicates the actual width of the material cross section; 、 Respectively represent the coordinates of the two end points of the circumference of the material; The corrected material circumferential width is calculated based on the second image pixel scale and the coordinates of the endpoints of the material circumferential width. The specific formula is: ； in, Indicates the circumferential width of the material after correction.

7. The intelligent detection method for cross-sectional characteristics of round hole profile of composite rod and wire according to claim 1, characterized in that: The material height is calculated based on the coordinates of the endpoints of the material circumference, specifically including: Determine the horizontal coordinate of the midpoint in the widening direction according to the coordinates of the endpoints of the widening circumference of the material; Determine the horizontal coordinate value range based on the set horizontal coordinate deviation and the horizontal coordinate of the midpoint; Determine the upper and lower intersection point numbers of the vertical line passing through each abscissa within the abscissa value range and the outer edge, and then determine the corresponding ordinate of each abscissa within the abscissa value range and the midpoint of the width direction; The coordinates of the upper groove bottom arc center point corresponding to each midpoint are determined according to the vertical distance difference between the arc center point of the upper groove bottom of the roll pass and the actual midpoint in the width direction; the coordinates of the lower groove bottom arc center point corresponding to each midpoint are determined according to the vertical distance difference between the arc center point of the lower groove bottom of the roll pass and the actual midpoint in the width direction; Set the value range of the upper intersection point number and the lower intersection point number; Calculate the distance between each upper intersection point within the range of the upper intersection point sequence number and the corresponding upper groove bottom arc center point, and the distance between each lower intersection point within the range of the lower intersection point sequence number and the corresponding lower groove bottom arc center point, and form a distance set from all distances; Calculating the variance of each distance in the distance set; Select the middle point corresponding to the distance with the smallest variance as the center point of the material; Determine the upper and lower intersection points of the vertical line passing through the center point of the material and the outer edge, and calculate the material height based on the upper and lower intersection points.

8. The intelligent detection method for cross-sectional characteristics of round hole profile of composite rod and wire according to claim 1, characterized in that: Calculate the coating thickness at each pixel point on the core edge based on the edge image and the circumferential width of the material, including: Take the intersection point of the straight line where the circumference of the material is widened and the left side of the core edge and the outer edge as the first point of the core edge and the outer edge, and sort all the pixel points of the core edge and the outer edge respectively; The distance between the first point on the edge of the core and the first point on the edge of the outer layer is used as the second reference distance; Calculate the distance between each pixel point on the core edge and other pixel points on the outer edge; where other pixel points on the outer edge refer to pixel points other than the first pixel point on the outer edge; Find the minimum distance between each pixel point on the edge of the core and other pixels on the edge of the outer layer; If the minimum distance is less than the second reference distance, the minimum distance is the coating thickness of the corresponding pixel point at the edge of the core; if the minimum distance is greater than or equal to the second reference distance, the second reference distance is the coating thickness of the corresponding pixel point at the edge of the core.

9. The intelligent detection method for cross-sectional characteristics of round hole profile of composite rod and wire according to any one of claims 1 to 8, characterized in that: The detection method further includes calculating a material circumference evaluation index to evaluate the material circumference characteristics; wherein the material circumference evaluation index includes a relative reduction of the current pass, a relative widening of the current pass, a circumferential unevenness coefficient of the material circumference, a circumferential fluctuation coefficient of the material circumference, a local weakness index of the material circumference, ear size, a proportion of the coating detection area of ​​the current pass, a total area error rate, a core area error rate, a coating area error rate, and an extension coefficient of the current pass; Calculate the relative reduction of the current pass based on the material height and the billet height before rolling; Calculate the relative width of the current pass based on the circumferential width of the material and the width of the billet before rolling; According to the circumference length of each pixel point of the outer edge of the material, the maximum circumference length of the material, the minimum circumference length of the material, the mean circumference length of the material and the standard deviation of the circumference length of the material are calculated. The ratio of the maximum circumference length of the material to the minimum circumference length of the material is used as the circumferential unevenness coefficient of the material circumference. The ratio of the standard deviation of the circumference length of the material to the mean circumference length of the material is used as the circumferential fluctuation coefficient of the material circumference. The ratio of the minimum circumference length of the material to the mean circumference length of the material is used as the local weakness index of the material circumference. Calculate the ear size of the cross section of the material according to the circumferential width of the material and the theoretical design width; Calculate the proportion of the current pass's coating inspection area based on the coating inspection area and the total inspection area; The total area error rate is calculated based on the theoretical design total area and the total detection area. The core area error rate is calculated based on the theoretical design core area and the core detection area. The cladding area error rate is calculated based on the theoretical design cladding area and the cladding detection area. Calculate the elongation factor of the current pass based on the cross-sectional area of ​​the blank and the total inspection area.

10. The intelligent detection method for cross-sectional characteristics of round hole profile of composite rod and wire according to any one of claims 1 to 8, characterized in that: The detection method further includes calculating a material coating evaluation index to evaluate the material coating characteristics; wherein the material coating evaluation index includes a circumferential unevenness coefficient of the coating, a circumferential fluctuation coefficient of the coating, a local weakness index of the coating, a number of wave peaks, and a number of wave troughs; The maximum cladding thickness, minimum cladding thickness, mean cladding thickness and standard deviation of cladding thickness are calculated based on the cladding thickness of each pixel point at the edge of the core. The ratio of the maximum cladding thickness to the minimum cladding thickness is used as the circumferential unevenness coefficient of the cladding, the ratio of the standard deviation of the cladding thickness to the mean cladding thickness is used as the circumferential fluctuation coefficient of the cladding, and the ratio of the minimum cladding thickness to the mean cladding thickness is used as the local weakness index of the cladding. A cladding thickness curve is drawn according to the cladding thickness of each pixel point at the edge of the core, and a horizontal line is drawn in the cladding thickness curve. The number of peaks and the number of troughs are determined based on the horizontal line.

11. An intelligent detection system for cross-sectional characteristics of round hole material of composite rod and wire, characterized in that: The detection system comprises: An acquisition unit is used to acquire a calibration plate image and a material cross-sectional image; obtain an edge image according to the material cross-sectional image, wherein the edge image includes an outer layer edge and a core edge; A first calculation unit, configured to calculate a first image pixel scale according to the calibration plate image; A second calculation unit is used to calculate the circumference length of the material mold at each pixel point of the outer edge according to the edge image; A third calculation unit is used to calculate the circumferential width of the material mold and the coordinates of its endpoints based on the pixel scale of the first image and the horizontal leftmost pixel point and the horizontal rightmost pixel point of the outer edge; A fourth calculation unit is used to calculate the height of the material mold according to the coordinates of the endpoints of the circumferential width of the material mold; A fifth calculation unit is used to calculate the coating thickness of each pixel point at the edge of the core according to the edge image and the circumferential width of the material; The sixth calculation unit is configured to calculate a core detection area and a total detection area according to the edge image, and calculate a cladding detection area according to the core detection area and the total detection area.

Images