Bare copper wire manufacturing quality detection method

Abstract

The invention relates to the technical field of optical detection of metal wires, in particular to a bare copper wire manufacturing quality detection method which comprises the following steps: S1, dynamically inhibiting oxygen by polarized light; s2, tension collaborative exposure control; s3, interference fringe multi-angle acquisition: performing beam splitting in an orthogonal polarized light path to generate reference light and object light, and acquiring interference images at azimuth angles of 0 degree, 30 degrees and 60 degrees through three synchronous cameras; s4, defect and deformation collaborative analysis: reconstructing the three-dimensional topology of the surface of the copper wire, and extracting a continuous region with depth gradient overrun as a crack; real-time ovality is calculated through line laser section projection; s5, multi-parameter coupling decision making: combining the maximum crack depth, the total crack length and the ovality change rate to construct a risk index; and when the risk index continuously exceeds the early warning threshold value, triggering a control instruction. Through polarized light dynamic oxygen suppression, tension collaborative exposure control and multi-angle interference image acquisition, cracks and defects on the surface of the copper wire can be accurately detected, and the detection precision is improved.

Description

Technical Field

[0001] This invention relates to the field of optical inspection technology for metal wires, and more particularly to a method for inspecting the manufacturing quality of bare copper wires. Background Technology

[0002] In the continuous casting and rolling process of bare copper wire, surface microcrack detection, geometric deformation monitoring, and production stability control constitute the core links in quality assurance. Currently, the industry commonly uses optical imaging technology for online inspection, but this has the following inherent drawbacks:

[0003] Bare copper wires exposed to air easily form a cuprous oxide (Cu2O) film, whose reflectivity in the visible light band differs significantly from that of the substrate. Traditional machine vision methods (such as grayscale thresholding and edge detection algorithms) cannot distinguish between oxide spots and real cracks because:

[0004] Oxidation spots exhibit localized dark area characteristics similar to cracks;

[0005] Fluctuations in oxide film thickness cause nonlinear changes in reflected light intensity, leading to the failure of fixed threshold segmentation.

[0006] Furthermore, tension fluctuations on the production line (±5%-8% of the rated value) cause slight vibrations in the copper wire (amplitude 0.1-0.5mm), while the line speed in high-speed continuous rolling often exceeds 200m / min. The existing technical solutions contain a fundamental contradiction:

[0007] If short exposure is used to suppress blur (such as 1μs level), the lighting intensity needs to be greatly increased, which will cause the copper wire surface to be overexposed and cover up the cracks.

[0008] If motion compensation algorithms (such as digital image stabilization) are enabled, the computation latency cannot meet the requirements for real-time detection (>50ms / frame).

[0009] Therefore, there is an urgent need for a method to inspect the manufacturing quality of bare copper wires and solve the above problems. Summary of the Invention

[0010] To achieve the above objectives, the present invention provides a method for quality inspection of bare copper wire manufacturing, comprising the following steps:

[0011] S1: Dynamic oxygen suppression by polarized light:

[0012] A bare copper wire is illuminated by a symmetrically arranged ring light source, and a first polarizer is set in the light path of the light source to form a first linear polarization direction;

[0013] A rotatable second polarizer is placed in front of the lens of a high-speed industrial camera;

[0014] Real-time closed-loop adjustment: Based on the gray value of the reflected light from the copper wire surface, the second polarizer is dynamically rotated to stabilize the gray value within the non-oxidized reference range;

[0015] S2: Tension-coordinated exposure control:

[0016] Real-time acquisition of tension fluctuation data and copper wire speed on the production line;

[0017] Displacement calculation: The instantaneous displacement of the copper wire is calculated based on the instantaneous change in tension, the copper wire speed, the current exposure time, and the pre-calibrated elastic compensation coefficient.

[0018] Adaptive exposure: When the instantaneous displacement exceeds the resolution threshold of the vision system, the exposure time is shortened;

[0019] S3: Multi-angle acquisition of interference fringes:

[0020] In an orthogonally polarized optical path, a reference beam and an object beam are generated by beam splitting. Interference images are acquired by three synchronous cameras at azimuth angles of 0°, 30°, and 60°.

[0021] S4: Defect and Deformation Synergistic Analysis:

[0022] Reconstruct the three-dimensional topology of the copper wire surface and extract the continuous region with the depth gradient exceeding the limit as the crack;

[0023] Real-time ellipticity is calculated by projecting a line laser cross section.

[0024] S5: Multi-parameter coupled decision-making:

[0025] A risk index is constructed using the maximum depth of the combined crack, the total crack length, and the rate of change of ellipticity.

[0026] A control command is triggered when the risk index exceeds the warning threshold for consecutive periods.

[0027] Preferably, the real-time closed-loop regulation of S1 specifically includes:

[0028] The grayscale value of reflected light from the surface of the oxide-free copper wire was collected in advance as the lower limit of the reference.

[0029] A standard copper oxide wire sample was prepared, and its reflected light gray value was collected as the upper limit of the benchmark.

[0030] The grayscale value of the reflected light on the surface of the copper wire is monitored in real time. When it exceeds the upper limit of the reference, the rotating mechanism is driven to adjust the angle of the second polarizer until the grayscale value drops below the lower limit of the reference.

[0031] The angle adjustment step size of the rotating mechanism is dynamically set according to the grayscale deviation: the larger the deviation, the larger the step size.

[0032] Preferably, obtaining the pre-calibrated elastic compensation coefficient of S2 includes:

[0033] In an offline state, a stepped increasing tension is applied to the same batch of copper wires, with each tension level being stable for a duration longer than the natural vibration period of the copper wire.

[0034] The maximum vibration amplitude under various tension levels is measured using a non-contact displacement sensor.

[0035] Plotting the change in tension on the x-axis and the amplitude of vibration on the y-axis, a straight line passing through the origin is fitted, and its slope is the elastic compensation coefficient.

[0036] Preferably, the method for determining the visual system resolution threshold of S2 is as follows:

[0037] Obtain the camera pixel size and lens magnification, and calculate the actual physical size corresponding to a single pixel;

[0038] The resolution threshold is set to an integer multiple of the physical size of a single pixel, and this multiple is selected according to the detection accuracy requirements.

[0039] Preferably, the reconstructed three-dimensional topology of the copper wire surface in S4 specifically includes:

[0040] Perform a four-step phase-shifting unwrapping process on each viewpoint interferogram:

[0041] The reference light phase shifts of 0°, 90°, 180°, and 270° were applied sequentially, and images were acquired.

[0042] Calculate the wrap phase using the intensity relationship of four images;

[0043] Phase continuity is expanded along the copper line axis, and compensation of an integer multiple of 2π is provided when the phase jump of adjacent pixels exceeds π.

[0044] The phase maps unwrapped from the three perspectives are converted into height gradient maps, which are then fused using the Poisson equation to generate a 3D topology.

[0045] Preferably, the continuous region where the extraction depth gradient exceeds the limit in S4 is taken as the crack, including:

[0046] Calculate the local radius of curvature for each pixel in the 3D topology;

[0047] When a certain region simultaneously satisfies:

[0048] a: The depth is greater than the diameter of the copper wire by a set ratio;

[0049] b: The average radius of curvature is less than the set threshold;

[0050] c: The area of ​​the connected region exceeds the minimum crack size;

[0051] If the condition is found to be a crack, the set ratio and threshold are determined through fracture mechanics experiments.

[0052] Preferably, the real-time ellipticity calculation of S4 via line laser section projection includes:

[0053] Extract the center point set of the line laser beam at the cross section of the copper wire;

[0054] Iterative fitting of the ellipse equation based on the random sampling consensus algorithm:

[0055] Calculate the ellipse parameters by randomly selecting 5 points;

[0056] Count the number of points in the entire point set that satisfy the error threshold of the ellipse equation;

[0057] Repeat the iteration until the ellipse model with the most interior points is found;

[0058] The absolute deviation ratio between the minor axis and the major axis is calculated based on the optimal ellipse model.

[0059] Preferably, obtaining the ellipticity change rate of S5 includes:

[0060] Ellipticity data is continuously collected over a fixed time window.

[0061] Perform linear regression analysis on the data within the window, and use the slope of the regression line as the rate of change;

[0062] The length of the time window is adaptively adjusted according to the copper wire speed: the higher the speed, the shorter the window length.

[0063] Preferably, in the construction risk index of S5:

[0064] The warning threshold is dynamically adjusted according to the diameter of the copper wire:

[0065] A direct proportional relationship between diameter and threshold is established, and the proportionality coefficient is determined through a fracture criticality experiment;

[0066] The current threshold is calculated by acquiring the copper wire diameter data from the production line PLC in real time.

[0067] Preferably, the triggering logic of S5 for triggering the control command when the risk index continuously exceeds the warning threshold includes: recording the event when the risk index first exceeds the threshold;

[0068] When the index exceeds the threshold for three consecutive detection cycles, and the rate of change shows an upward trend:

[0069] A speed reduction command is sent to the production line, and the speed reduction magnitude is positively correlated with the magnitude of the index exceeding the standard.

[0070] Activate the audible and visual alarm and locate the defect.

[0071] The beneficial effects of this invention are:

[0072] 1. This invention utilizes "polarized light dynamic oxidation suppression" technology. Under the influence of polarized light, the angle of the second polarizer can be dynamically adjusted to monitor the grayscale value of the reflected light on the copper wire surface in real time, maintaining it within the oxidation-free reference range. When oxidation occurs on the copper wire surface, the system removes the effects of oxidation through real-time closed-loop adjustment, effectively avoiding the confusion between oxide spots and cracks. This dynamic adjustment can accurately distinguish between oxide spots and cracks, improving the accuracy of detection.

[0073] 2. This invention dynamically rotates the second polarizer and adjusts the grayscale value of the reflected light in real time, keeping it within the reference range of copper oxide-free conditions. This avoids changes in reflected light intensity caused by fluctuations in oxide film thickness, thus ensuring the stability of reflected light intensity and overcoming the segmentation failure problem of traditional methods under light intensity variations. This method greatly improves the stability and reliability of detection.

[0074] 3. The "tension-coordinated exposure control" technology of this invention collects real-time tension fluctuation data and copper wire speed from the production line, combines this with a pre-calibrated elastic compensation coefficient, calculates the instantaneous displacement of the copper wire, and adaptively adjusts the exposure time. When the displacement exceeds the resolution threshold of the vision system, the exposure time is automatically shortened, avoiding image blurring caused by excessive exposure time. Simultaneously, the dynamic exposure adjustment and automatic compensation system effectively suppresses the blurring effect caused by copper wire vibration, ensuring image clarity and accuracy.

[0075] 4. This invention avoids the latency problem of traditional motion compensation algorithms by employing "adaptive exposure" technology. When displacement is caused by copper wire vibration, the system can quickly adjust the exposure time without relying on computationally intensive motion compensation algorithms. By dynamically adjusting the exposure time, the system can respond to minute vibrations of the copper wire in real time while ensuring stable image quality.

[0076] 5. This invention utilizes "multi-parameter coupled decision-making" technology, combining multiple detection parameters such as crack depth, crack length, and ellipticity change rate to construct a risk index in real time and set early warning thresholds. When the risk index continuously exceeds the early warning threshold, the system can immediately trigger control commands to adjust the production line or issue an alarm, thereby effectively preventing further escalation of quality problems. The system's real-time feedback and control capabilities significantly improve the production line's quality assurance capabilities and response speed. Attached Figure Description

[0077] To more clearly illustrate the technical solutions in this invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, those skilled in the art can obtain other drawings based on these drawings without creative effort.

[0078] Figure 1This is a flowchart of the steps of the method of the present invention;

[0079] Figure 2 This is a flowchart of the steps for obtaining the ellipticity change rate in method S5 of the present invention.

[0080] Figure 3 This is a flowchart illustrating the triggering logic of the control command in the method of the present invention when the risk index continuously exceeds the warning threshold. Detailed Implementation

[0081] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. It should also be noted that, to make the embodiments more comprehensive, the following embodiments are the best and preferred embodiments, and those skilled in the art can use other alternative methods to implement some well-known technologies; moreover, the accompanying drawings are only for more specific description of the embodiments and are not intended to specifically limit the present invention.

[0082] Please see Figures 1-3 This invention provides a method for quality inspection in bare copper wire manufacturing. S1 During the bare copper wire production process, a cuprous oxide film easily forms on the copper wire surface, causing changes in light reflection characteristics. To accurately distinguish between oxide spots and cracks, this invention employs a symmetrically arranged ring light source. A first polarizer forms linearly polarized light from the light source's output path, and a rotatable second polarizer adjusts the polarization direction. Through real-time closed-loop adjustment, based on the grayscale value of the reflected light from the copper wire surface, the second polarizer is dynamically rotated to stabilize the grayscale value of the reflected light within the oxide-free reference range. This method effectively eliminates the influence of the oxide film on the intensity of reflected light, thereby avoiding misjudgments in detecting oxide spots using traditional methods and significantly improving the accuracy of distinguishing between cracks and oxide spots.

[0083] During the S2 production process, the copper wire experiences slight vibrations due to tension fluctuations, affecting image clarity. To address this, this invention collects real-time data on production line tension fluctuations and copper wire speed, calculating the instantaneous displacement of the copper wire. Adaptive exposure control is then implemented based on the instantaneous displacement of the copper wire using a pre-calibrated elastic compensation coefficient and the current exposure time. When the copper wire displacement exceeds the resolution threshold of the vision system, the exposure time is automatically shortened, thus avoiding image blurring caused by prolonged exposure. This adaptive exposure control effectively improves image clarity in high-speed production environments, ensuring the accuracy of real-time detection.

[0084] In S3, to more comprehensively capture defect information on the copper wire surface, this invention employs an orthogonal polarization optical path, generating reference and object beams through beam splitting, and using three synchronous cameras to acquire interference images from different azimuth angles of 0°, 30°, and 60°. This multi-angle acquisition technology can comprehensively analyze minute defects on the copper wire surface from multiple perspectives, improving the accuracy of defect identification and reducing information that may be missed by single-angle acquisition.

[0085] In S4, a 3D topological map of the copper wire surface is reconstructed, and continuous regions with excessive depth gradients are extracted and identified as cracks. Simultaneously, the real-time ellipticity of the cracks on the copper wire surface is calculated using line laser cross-sectional projection technology. This method combines information such as crack depth, length, and morphology, which helps to accurately identify minute cracks on the copper wire surface and reduces the possibility of missed detections.

[0086] In S5, this embodiment of the invention comprehensively considers parameters such as the maximum depth, total length, and ellipticity change rate of the crack to construct a risk index. By monitoring these parameters in real time, when the risk index continuously exceeds a preset threshold, the system automatically triggers control commands to adjust the production line or issue an alarm. This decision-making mechanism enables early warning, prevents further expansion of quality problems, and ensures the stability of the production process and the quality of the copper wire.

[0087] This method solves problems such as difficulty in distinguishing between oxide spots and cracks, ambiguity caused by vibration, and insufficient exposure control. Through dynamic oxidation suppression with polarized light, tension-coordinated exposure control, and multi-angle interferometric image acquisition, this invention can accurately and in real-time detect cracks and defects on the surface of copper wires, significantly improving detection accuracy and real-time performance. Simultaneously, the multi-parameter coupled decision-making mechanism effectively prevents potential quality risks, ensures the stability of the production line, and ultimately improves the manufacturing quality of bare copper wires.

[0088] In one possible implementation, before starting the actual testing, it is necessary to first collect the gray value of the reflected light from the surface of the unoxidized copper wire and use it as the lower limit of the reference, that is, the standard value of the reflected light from the surface of the unoxidized copper wire.

[0089] Then, a standard copper oxide wire sample was prepared, and its reflected light grayscale value was collected and used as a reference upper limit. This process helped determine the range of variation in reflected light grayscale when oxidation occurs on the copper wire surface, ensuring that the detection system can correctly identify oxidized areas and normal areas.

[0090] Next, real-time monitoring and dynamic adjustment: During the production process, the grayscale value of reflected light on the copper wire surface is monitored in real time. This data is acquired using a high-precision industrial camera and transmitted to the system for analysis. When the grayscale value of reflected light exceeds the upper limit of the reference value, it indicates that oxidation may have occurred on the copper wire surface, and the system immediately activates the adjustment mechanism.

[0091] The system adjusts the angle of the second polarizer by driving a rotating mechanism, thereby changing the polarization direction of the light source and adjusting the grayscale of the reflected light from the copper wire surface. By continuously rotating the polarizer, the grayscale value is reduced to below the lower limit of the reference value, indicating that the surface has returned to an oxidation-free state.

[0092] The adjustment step size of the rotating mechanism is dynamically set based on the grayscale deviation. That is, if the current grayscale value of the reflected light differs significantly from the reference value, the system will increase the angle adjustment step size of the polarizer to respond quickly and reduce the grayscale of the reflected light to below the reference lower limit as quickly as possible. Conversely, if the grayscale value is close to the reference lower limit, the system will decrease the step size to avoid over-adjustment, thereby achieving the target grayscale value more accurately.

[0093] The dynamic closed-loop adjustment method can quickly adjust the polarization direction of the light source when oxidation occurs on the copper wire surface, ensuring that the gray value of the reflected light from the copper wire surface is always within the standard range of oxidation-free areas, avoiding interference from oxidation regions and improving detection accuracy. The rotation step size is automatically adjusted according to the gray value deviation, allowing the system to react quickly to large changes in reflected light gray value and make more precise adjustments for minor changes, thus optimizing the efficiency and accuracy of the adjustment process.

[0094] In one possible implementation, a series of physical tests are first required on the copper wire before obtaining the elasticity compensation coefficient. In an offline state, copper wires from the same batch are fixed and subjected to progressively increasing tensions. These tensions are increased gradually, and the duration of each tension level must be longer than the inherent vibration period of the copper wire. The purpose of this is to allow the copper wire to reach a steady state at each tension stage, ensuring that the vibration and response of the copper wire are in a measurable and stable state. By maintaining a steady-state time longer than the inherent vibration period, errors caused by instantaneous changes are eliminated, ensuring the accuracy and reliability of the test data.

[0095] After each tension level is applied, the maximum vibration amplitude of the copper wire is measured using a non-contact displacement sensor. This non-contact measurement method avoids the influence of mechanical interference and additional friction, and can accurately capture the true vibration characteristics of the copper wire at each tension stage. By measuring the maximum vibration amplitude under different tension conditions, the dynamic response characteristics of the copper wire are obtained, providing data support for subsequent calculations of the elastic compensation coefficient.

[0096] Next, the collected vibration data is processed. The horizontal axis represents the tension change, and the vertical axis represents the measured vibration amplitude. These data points are used for linear fitting to obtain a straight line passing through the origin. According to the linear regression formula, the slope of the fitting result is the elastic compensation coefficient. The elastic compensation coefficient reflects the degree of change in the vibration amplitude of the copper wire under different tensions, and has elastic response characteristics in a physical sense.

[0097] By applying incremental tension offline and combining it with the elastic compensation coefficient obtained through non-contact measurement, a scientific basis can be provided for real-time dynamic control of copper wire displacement and exposure time during production, significantly improving detection accuracy and production efficiency.

[0098] In one possible implementation, the camera's pixel size and lens magnification need to be obtained first. Pixel size is the physical size of each pixel on the camera's image sensor, typically measured in micrometers. Lens magnification refers to the proportion by which the lens magnifies the image of an object onto the sensor, reflecting the magnification capability of the camera's imaging system. Using these two parameters, the size of each pixel in the actual physical world is calculated. The calculation formula is as follows:

[0099] Single pixel physical size = pixel size / lens magnification; this calculation result shows the actual physical size represented by each pixel.

[0100] After calculating the physical size represented by each pixel, the next step is to set a resolution threshold based on the actual application scenario. The resolution threshold is the smallest scale at which the vision system can accurately identify and distinguish object details. To ensure the accuracy of quality inspection, the resolution threshold needs to be set as an integer multiple of the physical size of a single pixel. This multiple is determined based on the required detection accuracy.

[0101] The choice of multiple is based on the following factors:

[0102] Detection accuracy requirements: For example, if the requirement is to detect very small defects or changes, a smaller magnification should be selected to ensure that the image has sufficient detail resolution.

[0103] The size and detail of the object: For larger copper wires, the magnification should be increased appropriately, while for more precise inspection, a higher resolution may be required to ensure that the inspection system can capture minute defects.

[0104] Based on a set resolution threshold, the vision system processes the image, comparing the physical size of each pixel with the actual object size to determine if defects or anomalies exist. If the feature size of the object is below the resolution threshold, the system will be unable to recognize the detail, potentially leading to detection failure.

[0105] By reasonably calculating and setting the resolution threshold of the vision system, it is possible to optimize the detection accuracy and effect according to the requirements of the actual production environment while ensuring image clarity, thus providing reliable technical support for the quality inspection of bare copper wires.

[0106] In one possible implementation, phase unwrapping processing is required for the interferometric images at each viewpoint before performing 3D surface reconstruction. This process is accomplished using a four-step phase-shifting method. The specific steps are as follows:

[0107] In the interferogram at each viewpoint, reference light phase shifts of 0°, 90°, 180°, and 270° are applied sequentially. Each phase shift corresponds to the acquisition of one image. Different phase shifts allow the intensity variations in the interferogram to reflect the phase information of the object's surface.

[0108] After each phase shift is applied, the system acquires four corresponding images, which contain intensity data about the surface information of the copper wire.

[0109] By calculating the intensity relationship of the four images, the wrapping phase is obtained. Wrapping phase refers to the phase wrapping effect caused by the physical limitations of the imaging system, where the directly measured phase value repeatedly jumps between 0 and 2π.

[0110] After completing the four-step phase shift, phase continuity needs to be expanded along the axial direction of the copper wire to ensure spatial consistency of the phase. The specific steps are as follows:

[0111] When the phase jump between adjacent pixels exceeds π, it indicates that there is a phase overlap phenomenon, which needs to be compensated. The specific compensation method is to compensate the portion of the phase jump that exceeds π by an integer multiple of 2π, thereby restoring its correct phase value.

[0112] This phase continuity unfolding and compensation is a key step, ensuring that the phase information is not disturbed, thus providing a reliable data foundation for generating accurate three-dimensional surface models.

[0113] After unwrapping the phase map, the next step is to convert the unwrapped phase map into a height gradient map. There is a one-to-one correspondence between phase and the height of an object's surface. Using this relationship, the phase information is converted into height data of the object's surface. This process is achieved by leveraging the linear relationship between phase and height, mapping the phase map to the surface's height gradient.

[0114] By employing a four-step phase-shifting unwrapping, phase expansion compensation, phase-to-height conversion, and Poisson equation fusion, the system not only ensures high precision and stability in the three-dimensional topology reconstruction of bare copper wire surfaces but also enhances its detection capabilities and adaptability, demonstrating significant technical advantages.

[0115] After converting the unwrapped phase map into a height gradient map, a complete 3D surface model needs to be generated using mathematical methods. This step uses the Poisson equation to fuse the height gradient maps, thereby reconstructing the 3D topology of the object's surface. The Poisson equation can handle the recovery of complete surface height information from gradient data, solving surface reconstruction problems caused by insufficient or discontinuous gradient data.

[0116] The Poisson equation here serves to transform the local information of the surface height (i.e., the height gradient) into a globally consistent three-dimensional surface, ultimately yielding a three-dimensional topological model of the copper wire surface.

[0117] In one possible implementation, in a 3D topology map, each pixel represents a specific location on the copper wire surface. To analyze the shape of surface defects, it is first necessary to calculate the local radius of curvature for each pixel. The radius of curvature is an important parameter describing surface curvature, reflecting the degree of bending of the surface morphology. Specific calculation methods typically use the spatial coordinates of neighboring pixels to solve for the quadratic curvature of the local surface; common methods include gradient methods or fitting methods. A small local radius of curvature indicates drastic surface changes in that region, potentially indicating the presence of cracks or defects.

[0118] To identify cracks, three conditions are needed to determine whether a certain area is cracked:

[0119] First, determine whether the depth of the area (i.e., the change in height of the surface relative to the copper wire reference plane) is greater than a certain predetermined proportion of the copper wire diameter. This proportion is calibrated through fracture mechanics experiments to ensure that the crack depth is large enough to affect the integrity of the copper wire. The choice of this predetermined proportion usually depends on the crack detection requirements in the actual application and is generally set to 0.1 to 0.5 times the copper wire diameter.

[0120] Secondly, it is determined whether the average radius of curvature of the region is less than a preset threshold. Since cracked areas are usually characterized by large surface curvature, and areas with small radii of curvature often indicate severe surface deformation, the threshold setting for the radius of curvature is an important reference in crack detection.

[0121] Finally, it is necessary to check whether the region has sufficient connectivity to form a complete crack morphology. The connected area must exceed the minimum crack size to be considered a valid crack. This minimum size is also determined experimentally, based on the copper wire manufacturing process and practical application.

[0122] To ensure the accuracy of crack detection, the setting of the scale, radius of curvature threshold, and minimum crack size needs to be calibrated through fracture mechanics experiments. Fracture mechanics experiments can determine reasonable thresholds and scales by simulating or actually testing the effects of cracks of different depths, curvatures, and areas on the mechanical properties of copper wire. Experimental data provides a scientific basis for setting these parameters, ensuring the reliability of crack assessment.

[0123] By comprehensively analyzing the depth, curvature, and connectivity of the three-dimensional topological map of the copper wire surface, and combining the set ratio and threshold calibrated by fracture mechanics experiments, efficient and accurate crack detection is achieved, which has significant technical advantages and application prospects.

[0124] In one possible implementation, during online laser scanning, a laser beam is projected onto the cross-section of a copper wire. The laser light reflected or scattered from the surface of the copper wire forms light stripes on the image sensor. The set of center points corresponding to these light stripes represents the geometric contour of the copper wire cross-section. The distribution of these center points represents the actual shape of the copper wire cross-section, which may be circular, elliptical, or other shapes.

[0125] Random Sample Consensus (RANSAC) is an algorithm commonly used to handle noise and outliers in data. The main idea of ​​RANSAC is to randomly select a subset of points from the data; in this embodiment, these points represent a correct model. The algorithm then fits the model and calculates the error between all data points and the model to determine its validity. Its main steps are as follows:

[0126] Five points are randomly selected from the set of center points of the light stripe. These five points will be used to fit the basic parameters of the ellipse, including the major axis, minor axis, and inclination angle. Using the geometric relationships of these five points, an equation of the ellipse is calculated. The standard equation of the ellipse is:

[0127]

[0128] Where a and b are the major axis and minor axis of the ellipse, respectively.

[0129] After calculating the equation of the ellipse, it is necessary to evaluate the degree of matching between the center points of all light stripes and the ellipse. Specifically, the distance from each point to the fitted ellipse is calculated and compared with a set error threshold. If the distance from a point to the ellipse is less than the error threshold, the point is considered to conform to the ellipse model and is called an "interior point".

[0130] The RANSAC algorithm repeatedly performs the above steps until it finds the elliptical model with the most interior points. Each iteration selects a new elliptical model based on different random point choices. Through multiple iterations, the elliptical model with the most interior points is ultimately selected as the optimal model.

[0131] Using the optimal ellipse model, calculate the major axis (a) and minor axis (b) of the ellipse, and determine the ratio of their absolute deviations, i.e.:

[0132]

[0133] This deviation ratio is used to characterize the ellipticity of the copper wire cross-section. A larger deviation indicates that the copper wire cross-section is close to an ellipse, suggesting a significant shape defect. Conversely, a smaller deviation indicates a more regular cross-section and better quality.

[0134] By combining line laser projection and the RANSAC algorithm, the accuracy of copper wire manufacturing quality inspection is improved, and it also has strong real-time performance and adaptability, showing broad application prospects.

[0135] In one possible implementation, during the copper wire production process, a laser scanning system of a certain frequency is used to continuously measure the ellipticity of the copper wire, recording the ellipticity value at each moment. This data is collected within a preset time window. The length of the time window is typically a fixed value, and the ellipticity changes within this period are analyzed as a dataset. In this embodiment of the invention, the time window length is T, during which the system records multiple ellipticity values ​​based on changes in the cross-section of the copper wire.

[0136] After collecting a certain amount of ellipticity data, linear regression analysis is performed. The purpose of linear regression analysis is to fit the data points using the least squares method and find a regression line. The slope of this regression line represents the rate of change of ellipticity over time, i.e., the rate of change of ellipticity. Specifically, in this embodiment of the invention, the collected ellipticity data points are (t1, e1), (t2, e2), ..., (t... n e n ), where t is the time point and e is the corresponding ellipticity value. The slope m of the regression line represents the trend of ellipticity change:

[0137]

[0138] The slope represents the rate of change of ellipticity; the larger the slope, the faster the ellipticity changes, and the smaller the slope, the slower the change.

[0139] Since the production speed of copper wire may vary under different conditions, a fixed time window length may not be able to reflect the impact of speed changes on ellipticity monitoring in real time. Therefore, the time window length needs to be dynamically adjusted according to the copper wire production speed. Specifically, when the copper wire speed is high, the time window length should be shortened to ensure that the number of data points within each time window is suitable for the rapidly changing production environment; conversely, when the copper wire speed is low, the time window length should be appropriately extended to ensure sufficient data for regression analysis. In this way, by adaptively adjusting the time window length, the matching of data acquisition with the copper wire speed can be ensured, improving the accuracy and real-time performance of the detection.

[0140] By combining linear regression analysis and adaptive adjustment of the time window length, efficient and accurate real-time monitoring can be achieved, adapting to different production environments and providing strong support for the quality control of copper wire.

[0141] In one possible implementation, the diameter of the copper wire is closely related to its mechanical properties, shape stability, and other quality characteristics during the copper wire production process. Experimental studies, particularly fracture criticality experiments, have established a direct proportional relationship between the copper wire diameter and the risk warning threshold. This relationship indicates that the larger the diameter of the copper wire, the higher the requirement for quality stability, and the higher the warning threshold should be. This is because larger diameter copper wires may experience more stress or have more morphological defects during production, thus requiring a higher risk threshold to detect quality problems early.

[0142] To accurately determine the proportionality coefficient between copper wire diameter and risk threshold, a fracture criticality experiment was conducted. The experiment tested the fracture behavior of copper wires of different diameters under varying stress conditions, recording the critical state of the copper wire before breakage. This data was then compared with the diameter of the copper wire to determine the appropriate proportionality coefficient. This coefficient quantifies the relationship between the copper wire diameter and the risk threshold, allowing for precise control of the risk threshold adjustment based on the actual diameter of the copper wire.

[0143] During production, the diameter of the copper wire is typically monitored and recorded in real time by a PLC (Programmable Logic Controller) on the production line. This controller accurately collects the current diameter data of the copper wire and transmits it to the quality inspection system in real time. By combining this diameter data with the actual diameter data, the system can adjust risk warning thresholds in real time to respond promptly to quality changes on the production line.

[0144] Based on real-time acquired copper wire diameter data, the current risk warning threshold is calculated using established proportional relationships and coefficients. This threshold is dynamically adjusted, allowing the system to set different quality monitoring standards for copper wires of different diameters. If the diameter is larger, the warning threshold will increase, and vice versa. This adjustment accurately reflects changes in risk during the production process, ensuring the system can issue timely quality warnings.

[0145] By dynamically adjusting the early warning threshold and monitoring the copper wire diameter in real time, more accurate and flexible quality risk assessment and early warning can be achieved, effectively improving the quality control level in the copper wire manufacturing process and providing strong technical support for quality management in the production process.

[0146] In one possible implementation, during the production process, the quality inspection system calculates the risk index of the copper wire in real time and compares it with a preset risk warning threshold. When the risk index is detected to exceed the warning threshold for the first time, the system immediately records this event. The core function of this step is to provide a data foundation for subsequent quality analysis and problem tracing, ensuring that when quality problems occur, they can be traced back to the specific time and environmental conditions in which the problem occurred.

[0147] After the threshold is exceeded for the first time, the system will continue to monitor the risk index for several subsequent monitoring cycles. If the index exceeds the threshold for three consecutive monitoring cycles, and these index values ​​show an upward trend, indicating that the risk index is gradually increasing, it suggests that the problem in the production process may be escalating. At this point, the system considers the risk situation to have reached a severe level and triggers further control measures.

[0148] When the above conditions are met, the system will send a speed-down command to the production line based on the extent to which the risk index exceeds the limit. The speed-down is directly proportional to the extent to which the risk index exceeds the limit; that is, the greater the exceedance, the greater the speed-down. The purpose of this measure is to reduce the production pressure on copper wire by slowing down the production speed, thereby reducing quality problems caused by excessive speed and ensuring that the quality of copper wire products remains within a controllable range.

[0149] To promptly alert production operators, the system activates audible and visual alarms, emitting a clear warning signal. Simultaneously, the system utilizes real-time monitoring data analysis to pinpoint the exact location or area where the quality issue has occurred. This helps operators respond quickly and take appropriate measures to prevent the quality problem from escalating.

[0150] Through sophisticated trigger control command logic and an adaptive risk management mechanism, quality risks can be monitored in real time during copper wire production, respond quickly and make effective adjustments to ensure production stability and copper wire quality pass rate, while improving production efficiency and automation level.

[0151] The following examples will illustrate this in detail:

[0152] This embodiment applies to a copper wire manufacturing plant that produces copper wire of different specifications using an automated production line. The main equipment on the production line includes a wire drawing machine, a stretching machine, and an annealing furnace. Key production parameters include the diameter of the copper wire, production speed, and temperature. In traditional production processes, quality problems with copper wire are often caused by fluctuations in production parameters. Manual intervention often suffers from response lag and cannot precisely control all products. This invention aims to improve the stability of the production process by calculating the risk index in real time and making automatic adjustments.

[0153] The core of this invention is to calculate a risk index based on the diameter of the copper wire and the production speed, and to use this index to assess potential quality problems during the production process.

[0154] The risk index R is calculated using the following formula:

[0155]

[0156] in:

[0157] C dCurrent copper wire diameter (unit: mm). C threshold : Standard copper wire diameter (unit: mm). In this embodiment of the invention, the standard copper wire diameter is 2.5 mm. V s : Production speed (unit: m / min). ΔT: Temperature fluctuation coefficient. In this embodiment of the invention, the temperature fluctuation range is 0.05 to 0.1, and the value is 0.07.

[0158] Standard copper wire diameter: C threshold =2.5mm;

[0159] Production speed: V s The value range is 10–100 m / min, and should be adjusted according to the specific conditions of the production line.

[0160] Temperature fluctuation coefficient: ΔT = 0.07. In this embodiment, this value is fixed under normal production conditions.

[0161] In this embodiment, the collected data is as follows:

[0162] Copper wire diameter: C d =2.7mm;

[0163] Production speed: V s =50m / min

[0164] Temperature fluctuation coefficient: ΔT = 0.07;

[0165] Based on the above formula, the risk index R is calculated as follows:

[0166]

[0167] At this point, the risk index R = 57.78, which exceeds the set risk threshold (the set threshold is 50), indicating that there is a high quality risk in the current production process.

[0168] The system will automatically determine whether to make adjustments based on the risk index value. In this embodiment, if the risk index R exceeds the threshold of 50, the system will trigger an automatic adjustment mechanism to reduce the speed.

[0169] For example, when the risk index R = 57.78, the control system will issue a command to reduce the production speed V. s The speed is reduced to 40 m / min. This adjustment is immediately fed back to the PLC control system to control the operating speed of the production equipment.

[0170] Adjusted production speed V new Calculated using the following formula:

[0171] V new =V s -ΔV;

[0172] in:

[0173] V s Current production speed (50m / min).

[0174] ΔV: The speed adjustment range calculated based on the risk index. In this embodiment of the invention, for every unit increase in the risk index R, the production speed needs to be reduced by 0.1 m / min.

[0175] For a risk index R = 57.78, the adjustment range is...

[0176] ΔV=0.1×(R-50)=0.1×(57.78-50)=0.778m / min;

[0177] Therefore, the new production speed is:

[0178] V new =50-0.778=49.22m / min.

[0179] The adjusted production speed is 49.22 m / min, slightly lower than the original 50 m / min. The system ensures that the copper wire production process remains within an acceptable quality range by monitoring data such as diameter, speed, and temperature in real time, thus avoiding potential quality fluctuations.

[0180] To demonstrate the effectiveness of this invention, the following comparison is made:

[0181] In traditional copper wire production, quality control relies primarily on manual observation and inspection. This method is not only slow to respond but also prone to missing quality issues caused by sudden changes. For example, when the copper wire diameter suddenly deviates from the standard value, manual intervention is often delayed, leading to a large number of defective products during production.

[0182] Compared to traditional methods, this invention enables real-time detection and rapid response to quality fluctuations during production through automated monitoring and adjustment. For example, when the copper wire diameter exceeds the standard range, this invention can quickly assess the quality risk using a risk index model and automatically adjust production parameters, whereas traditional methods might take several hours to detect the problem.

[0183] The experiment compared the quality control effects of a production line using the system of this invention with those of a traditional production line. Experimental data showed that after introducing the automatic adjustment mechanism of this invention, the diameter deviation of the copper wire was reduced by approximately 30%, the fluctuation range of production speed was reduced by 25%, and the product qualification rate increased by 15%.

[0184] Comparative analysis shows that the present invention has significant advantages over traditional methods, effectively improving the quality and efficiency of copper wire production. Furthermore, the system achieves fully automated operation, reducing the impact of human factors.

[0185] This invention encompasses any substitutions, modifications, equivalent methods, and solutions made within the spirit and scope of this invention. To provide the public with a thorough understanding of this invention, specific details are described in detail in the following preferred embodiments; however, those skilled in the art will fully understand the invention even without these details. Furthermore, to avoid unnecessary misunderstanding of the nature of this invention, well-known methods, processes, procedures, components, and circuits are not described in detail.

[0186] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for quality inspection in bare copper wire manufacturing, characterized in that, Includes the following steps: S1: Dynamic oxygen suppression by polarized light: A bare copper wire is illuminated by a symmetrically arranged ring light source, and a first polarizer is set in the light path of the light source to form a first linear polarization direction; A rotatable second polarizer is placed in front of the lens of a high-speed industrial camera; Real-time closed-loop adjustment: Based on the gray value of the reflected light from the copper wire surface, the second polarizer is dynamically rotated to stabilize the gray value within the non-oxidized reference range; S2: Tension-coordinated exposure control: Real-time acquisition of tension fluctuation data and copper wire speed on the production line; Displacement calculation: The instantaneous displacement of the copper wire is calculated based on the instantaneous change in tension, the copper wire speed, the current exposure time, and the pre-calibrated elastic compensation coefficient. Adaptive exposure: When the instantaneous displacement exceeds the resolution threshold of the vision system, the exposure time is shortened; S3: Multi-angle acquisition of interference fringes: In an orthogonally polarized optical path, a reference beam and an object beam are generated by beam splitting. Interference images are acquired by three synchronous cameras at azimuth angles of 0°, 30°, and 60°. S4: Defect and Deformation Synergistic Analysis: Reconstruct the three-dimensional topology of the copper wire surface and extract the continuous region with the depth gradient exceeding the limit as the crack; Real-time ellipticity is calculated by projecting a line laser cross section. S5: Multi-parameter coupled decision-making: A risk index is constructed using the maximum depth of the combined crack, the total crack length, and the rate of change of ellipticity. A control command is triggered when the risk index exceeds the warning threshold for consecutive periods.

2. The method for quality inspection of bare copper wire manufacturing according to claim 1, characterized in that, The real-time closed-loop regulation of S1 specifically includes: The grayscale value of reflected light from the surface of the oxide-free copper wire was collected in advance as the lower limit of the reference. A standard copper oxide wire sample was prepared, and its reflected light gray value was collected as the upper limit of the benchmark. The grayscale value of the reflected light on the surface of the copper wire is monitored in real time. When it exceeds the upper limit of the reference, the rotating mechanism is driven to adjust the angle of the second polarizer until the grayscale value drops below the lower limit of the reference. The angle adjustment step size of the rotating mechanism is dynamically set according to the grayscale deviation: the larger the deviation, the larger the step size.

3. The method for quality inspection of bare copper wire manufacturing according to claim 1, characterized in that, The acquisition of the pre-calibrated elastic compensation coefficient of S2 includes: In an offline state, a stepped increasing tension is applied to the same batch of copper wires, with each tension level being stable for a duration longer than the natural vibration period of the copper wire. The maximum vibration amplitude under various tension levels is measured using a non-contact displacement sensor. Plotting the change in tension on the x-axis and the amplitude of vibration on the y-axis, a straight line passing through the origin is fitted, and its slope is the elastic compensation coefficient.

4. The method for quality inspection of bare copper wire manufacturing according to claim 1, characterized in that, The method for determining the visual system resolution threshold of S2 is as follows: Obtain the camera pixel size and lens magnification, and calculate the actual physical size corresponding to a single pixel; The resolution threshold is set to an integer multiple of the physical size of a single pixel, and this multiple is selected according to the detection accuracy requirements.

5. The method for quality inspection of bare copper wire manufacturing according to claim 1, characterized in that, The reconstructed three-dimensional topology of the copper wire surface in S4 specifically includes: Perform a four-step phase-shifting unwrapping process on each viewpoint interferogram: The reference light phase shifts of 0°, 90°, 180°, and 270° were applied sequentially, and images were acquired. Calculate the wrap phase using the intensity relationship of four images; Phase continuity is expanded along the copper line axis, and compensation of an integer multiple of 2π is provided when the phase jump of adjacent pixels exceeds π. The phase maps unwrapped from the three perspectives are converted into height gradient maps, which are then fused using the Poisson equation to generate a 3D topology.

6. The method for quality inspection of bare copper wire manufacturing according to claim 1, characterized in that, S4 extracts continuous regions exceeding the depth gradient limit as cracks, including: Calculate the local radius of curvature for each pixel in the 3D topology; When a certain region simultaneously satisfies: a: The depth is greater than the diameter of the copper wire by a set ratio; b: The average radius of curvature is less than the set threshold; c: The area of ​​the connected region exceeds the minimum crack size; If the condition is found to be a crack, the set ratio and threshold are determined through fracture mechanics experiments.

7. The method for quality inspection of bare copper wire manufacturing according to claim 1, characterized in that, S4's real-time ellipticity calculation via line laser section projection includes: Extract the center point set of the line laser beam at the cross section of the copper wire; Iterative fitting of the ellipse equation based on the random sampling consensus algorithm: Calculate the ellipse parameters by randomly selecting 5 points; Count the number of points in the entire point set that satisfy the error threshold of the ellipse equation; Repeat the iteration until the ellipse model with the most interior points is found; The absolute deviation ratio between the minor axis and the major axis is calculated based on the optimal ellipse model.

8. The method for quality inspection of bare copper wire manufacturing according to claim 1, characterized in that, The ellipticity change rate of S5 is obtained by: Ellipticity data is continuously collected over a fixed time window. Perform linear regression analysis on the data within the window, and use the slope of the regression line as the rate of change; The length of the time window is adaptively adjusted according to the copper wire speed: the higher the speed, the shorter the window length.

9. The method for quality inspection of bare copper wire manufacturing according to claim 1, characterized in that, In the S5 construction risk index: The warning threshold is dynamically adjusted according to the diameter of the copper wire: A direct proportional relationship between diameter and threshold is established, and the proportionality coefficient is determined through a fracture criticality experiment; The current threshold is calculated by acquiring the copper wire diameter data from the production line PLC in real time.

10. A method for quality inspection of bare copper wire manufacturing according to claim 1, characterized in that, The triggering logic for S5 to issue control commands when the risk index continuously exceeds the warning threshold includes: Record the event when the risk index first exceeds the threshold; When the index exceeds the threshold for three consecutive detection cycles, and the rate of change shows an upward trend: A speed reduction command is sent to the production line, and the speed reduction magnitude is positively correlated with the magnitude of the index exceeding the standard. Activate the audible and visual alarm and locate the defect.

Images