Metal wire data analysis method and system

Abstract

The invention relates to the technical field of data processing, in particular to a metal wire data analysis method and system. The method comprises the following steps: acquiring wire interface data of composite metal; acquiring a wire rod interface characteristic matrix based on the wire rod interface data; constructing a wire rod interface characteristic model according to the wire rod interface characteristic matrix; differential signal conversion processing is carried out on the wire rod interface characteristic model, and wire rod interface change characteristics are extracted based on a differential signal conversion result; constructing a wire rod interface defect initial indication diagram according to the wire rod interface change characteristics; performing multi-scale decomposition processing on the basis of the wire rod interface defect initial indication graph to obtain interface defect feature data; and identifying wire interface defect type features according to the interface defect feature data. According to the method, a closed-loop optimization mechanism of defect characteristics and production process parameters is established, and the interface quality and reliability of the high-performance composite metal wire are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a metal wire material data analysis method and system. Background Art

[0002] Metal wire is a slender metal material with a relatively small cross-sectional area and a long length, typically in a linear form. It can be made from a single metal (such as copper, aluminum, iron) or an alloy (such as stainless steel, nickel-titanium alloy). The diameter of a metal wire is typically less than or equal to 100 mm. Bimetallic composite wire is a composite wire composed of two different metals or alloys, with the core and outer layer each composed of a different metal material. This material combines the characteristics of both the core and outer layer materials, saving high-value metals while also compensating for the performance deficiencies of a single material. In the production of bimetallic composite wire (such as copper-clad aluminum wire), complex defects (such as delamination, pores, and inclusions) occur at the interface. These defects typically do not occur in isolation, but rather as a combination of multiple defects, resulting in highly coupled signals. This renders single-defect recognition algorithms ineffective, requiring the simultaneous modeling of complex relationships between multiple defect features, which is extremely challenging to label. Diffusion layers or intermediate layers often exist at the interface between copper and aluminum, resulting in blurred boundaries or even a gradual transition in imaging or spectral analysis. Traditional algorithms based on edge recognition and texture differences have difficulty in clearly extracting defect contours or locating defect areas. Summary of the Invention

[0003] Based on this, it is necessary for the present invention to provide a metal wire data analysis method and system to solve at least one of the above technical problems.

[0004] To achieve the above object, a metal wire data analysis method comprises the following steps:

[0005] Step S1: obtaining wire interface data of the composite metal; collecting a wire interface feature matrix based on the wire interface data; and constructing a wire interface feature model according to the wire interface feature matrix;

[0006] Step S2: performing differential signal conversion processing on the wire interface feature model, extracting wire interface change characteristics based on the differential signal conversion results; and constructing a preliminary wire interface defect indication map based on the wire interface change characteristics;

[0007] Step S3: performing multi-scale decomposition processing based on the preliminary indication map of the wire interface defect to obtain interface defect characteristic data; identifying the wire interface defect type characteristics based on the interface defect characteristic data; matching and correcting the wire interface defect type characteristics with a preset physical constraint model to generate accurate defect characterization data;

[0008] Step S4: establishing a wire interface defect distribution map based on the precise defect characterization data; calculating the wire interface defect severity based on the wire interface defect distribution map; and analyzing wire production process-related data based on the wire interface defect severity.

[0009] Step S5: Generate wire manufacturing process optimization data based on the wire production process associated data; apply the wire manufacturing process optimization data to adjust wire production process parameters, and verify the wire interface quality to obtain wire interface performance evaluation data.

[0010] The present invention further provides a metal wire data analysis system for executing the metal wire data analysis method described above, the metal wire data analysis system comprising:

[0011] The data acquisition and modeling module is used to obtain the wire interface data of the composite metal; collect the wire interface feature matrix based on the wire interface data; and construct the wire interface feature model based on the wire interface feature matrix;

[0012] The differential signal processing module is used to perform differential signal conversion processing on the wire interface feature model, extract the wire interface change characteristics based on the differential signal conversion results, and construct a preliminary indication map of wire interface defects based on the wire interface change characteristics;

[0013] The multi-scale decomposition module is used to perform multi-scale decomposition processing based on the preliminary indication map of the wire interface defect to obtain interface defect characteristic data; identify the type characteristics of the wire interface defect based on the interface defect characteristic data; match and correct the wire interface defect type characteristics with the preset physical constraint model to generate accurate defect characterization data;

[0014] The defect distribution module is used to establish a wire interface defect distribution map based on the precise defect characterization data; calculate the wire interface defect severity based on the wire interface defect distribution map; and analyze the wire production process related data based on the wire interface defect severity;

[0015] The quality verification module is used to generate wire manufacturing process optimization data based on wire production process related data; apply the wire manufacturing process optimization data to the wire production process parameter adjustment, and verify the wire interface quality to obtain wire interface performance evaluation data.

[0016] This method significantly improves the accuracy of the production process and the stability of wire quality by comprehensively analyzing and optimizing interface defects during metal wire production. First, by establishing a library of physical constraint models for wire interface defects and comparing them with parameterized defect characteristics, it enables precise identification and characterization of different defect types, helping to identify the mechanisms of each defect and its impact on wire quality. Each defect type can be finely classified based on its characteristics, and corresponding optimization strategies can be implemented for each defect, effectively reducing product quality fluctuations caused by defects. Second, by combining spatial distribution data of wire interface defects, it is possible to accurately map the defect distribution, providing an intuitive basis for subsequent quality assessment. This distribution map not only reflects the occurrence patterns of various defects but also reveals the changing trends of defects under different production process parameters. This combination of spatial positioning and distribution analysis facilitates more accurate assessment of wire quality levels, both overall and in local areas, providing early warning of potential quality issues that may arise during the production process. During the quality assessment process, this method, through correlation analysis between defect types and production process parameters, provides a deep understanding of the direct impact of process parameter changes on wire interface quality. This analysis clearly identifies the process parameters that are key drivers of specific defects, providing a theoretical basis for subsequent process adjustments. More importantly, the results can be used to optimize production process parameters, avoiding new defects caused by over-adjustment or inappropriate adjustments, thereby improving overall production efficiency and consistent product quality. Furthermore, this method demonstrates a high degree of flexibility and scientific validity in process parameter optimization. By precisely identifying key process parameters that impact interface quality and optimizing these parameters based on these parameters, it effectively reduces defect types caused by inappropriate process parameters, particularly defects such as delamination and porosity that significantly impact wire material performance. This process not only improves production efficiency but also reduces resource waste and production costs. After the actual process parameter adjustment plan is validated through simulation, the adjustment effect can be predicted in advance, providing more reliable guidance for the production line. This adjustment verification based on predicted data avoids the uncertainty associated with blind adjustments and further reduces the cost of the experimental phase. By converting the optimized process parameters into production equipment control instructions, the optimization plan can be more accurately executed during production, and process parameters can be fine-tuned in real time to ensure optimal operation at every stage. By adjusting production process parameters in real time and collecting relevant process data, this method enables precise control of various process parameters during the production process. The real-time data feedback mechanism not only ensures continuous optimization of production parameters but also provides detailed process data support for quality control. In this process, precise process adjustments and data collection provide a reliable basis for subsequent quality testing and enable rapid response to abnormalities in the production process.Finally, combined with the test results of interface quality, this method realizes the quantitative evaluation of the process optimization effect. By detecting the type, quantity, size and distribution of defects on the surface and inside of the wire, the improvement effect of process optimization on the quality of the wire can be comprehensively evaluated. This process ensures whether the various process adjustments in the production process meet the expected goals and provides a feedback mechanism for subsequent process adjustments. This feedback mechanism can not only help the production process team to continuously optimize the process, but also improve the quality standards of wire production through regular evaluation. In summary, this method accurately analyzes the generation mechanism of wire interface defects, combines real-time data in the production process, deeply explores the influence of process parameters on defects, and through scientific process optimization and real-time adjustment, makes every link of wire production operate in the best state. Through this series of interlocking optimization processes, the overall quality of wire products can be greatly improved, and unqualified products caused by defects can be reduced, thereby improving production efficiency and reducing production costs, which has significant practical application value. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Other features, objects and advantages of the present invention will become more apparent upon reading the detailed description of non-limiting embodiments thereof made with reference to the following drawings:

[0018] Figure 1 This is a schematic flow chart of the steps of a metal wire material data analysis method according to the present invention;

[0019] Figure 2 for Figure 1 Detailed step flow diagram of step S1;

[0020] Figure 3 for Figure 1 Detailed step flow chart of step S2 in FIG. DETAILED DESCRIPTION

[0021] The following is a clear and complete description of the technical method of the present invention in conjunction with the accompanying drawings. It is obvious that the embodiments described are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making any creative efforts are within the scope of protection of the present invention.

[0022] In addition, the accompanying drawings are merely schematic illustrations of the present invention and are not necessarily drawn to scale. Identical reference numerals in the figures denote identical or similar parts, and thus repetitive descriptions thereof will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities that do not necessarily correspond to physically or logically separate entities. These functional entities may be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor and / or microcontroller approaches.

[0023] It should be understood that although the terms "first," "second," and the like may be used herein to describe various elements, these elements should not be limited by these terms. These terms are used solely to distinguish one element from another. For example, a first element may be referred to as a second element, and similarly, a second element may be referred to as a first element, without departing from the scope of the exemplary embodiments. The term "and / or" as used herein includes any and all combinations of one or more of the listed associated items.

[0024] To achieve this, please refer to Figures 1 to 3 The present invention provides a metal wire material data analysis method, the method comprising the following steps:

[0025] Step S1: obtaining wire interface data of the composite metal; collecting a wire interface feature matrix based on the wire interface data; and constructing a wire interface feature model according to the wire interface feature matrix;

[0026] When performing data analysis on a composite metal wire, an embodiment of the present invention first obtains high-resolution image data of the wire cross section by combining a scanning electron microscope (SEM) with a focused ion beam (FIB) technique, collects microscopic interface information between different metal layers inside the wire, and the image data obtained is the wire interface data. Next, a grayscale gradient detection method based on image processing is used to extract interface contour features, and the interface material properties are encoded by combining a texture direction matrix (such as a grayscale co-occurrence matrix) and multi-channel color distribution statistics to construct a wire interface feature matrix. The feature matrix includes but is not limited to multidimensional feature parameters such as interface roughness, inclusion distribution, fusion boundary continuity, and metal particle orientation. Furthermore, the principal component analysis (PCA) algorithm is used to reduce the dimensionality and retain the main change dimensions to establish a wire interface feature model. The model expresses the spatial structure and material property change pattern of interfaces in different regions in a vectorized form, providing high-quality input data for subsequent defect identification.

[0027] Step S2: performing differential signal conversion processing on the wire interface feature model, extracting wire interface change characteristics based on the differential signal conversion results; and constructing a preliminary wire interface defect indication map based on the wire interface change characteristics;

[0028] After obtaining the wire interface feature model, the embodiment of the present invention adopts a differential signal conversion processing method that combines time domain difference and space domain difference to enhance the sensitivity of interface structure changes, wherein time domain difference is used to analyze the microstructural changes of the same area in the continuous production sample over time, and space domain difference is used to characterize the non-uniform distribution of the interface morphology of different wire cross sections at the same time. The signal data obtained after differential processing presents a "response curve" form of interface feature changes. By performing envelope extraction and peak density statistics on these differential responses, the interface mutation points and their degree of change are identified, thereby extracting the wire interface change characteristics, such as the location of the starting point of microcracks, the separation trend of the metal layer, etc. Subsequently, a preliminary defect indicator map is constructed based on these change characteristics. The map indicates the areas in the wire cross section that may have defects in the form of a pseudo-color heat map. The heat value indicates the severity of the change in the potential defect area, forming the basis for the first step of defect manifestation.

[0029] Step S3: performing multi-scale decomposition processing based on the preliminary indication map of the wire interface defect to obtain interface defect characteristic data; identifying the wire interface defect type characteristics based on the interface defect characteristic data; matching and correcting the wire interface defect type characteristics with a preset physical constraint model to generate accurate defect characterization data;

[0030] The embodiment of the present invention performs multi-scale decomposition processing on the generated preliminary indication map of wire interface defects, and uses the wavelet packet decomposition algorithm to perform multi-scale spatial feature decomposition on each potential defect area in the image to separate local detail features and global trend features. At each scale, defect shape features (such as ellipticity, angularity), spatial distribution features (such as directional consistency, concentration) and material response features (such as absorptivity change, density gradient) are extracted respectively. An interface defect feature data set is formed based on the above features. Further, the support vector machine (SVM) classifier is combined with the expert experience sample library to identify defect type features, such as specific defect categories such as poor fusion, foreign body inclusion, crack initiation, and interlayer debonding. On this basis, these defect features are constrained and corrected using a preset metal physical constraint model. The model is established based on parameters such as the thermal expansion coefficient, Young's modulus, and stress concentration coefficient of the metal, so that the detected defect type has physical rationality from the perspective of thermodynamics and material mechanics. After correction, accurate defect characterization data is generated to ensure the accuracy of subsequent evaluation.

[0031] Step S4: establishing a wire interface defect distribution map based on the precise defect characterization data; calculating the wire interface defect severity based on the wire interface defect distribution map; and analyzing wire production process-related data based on the wire interface defect severity.

[0032] The embodiment of the present invention maps the precise defect characterization data to a three-dimensional wire model to construct a wire interface defect distribution map. The map uses voxel representation structure, and each voxel records the defect type, location coordinates, and impact degree. A statistical method based on volume weight is used to analyze the defect map, and the defect severity score of each section of wire is calculated. The score is calculated based on three dimensions: defect volume density, stress concentration area superposition, and defect distribution symmetry. Subsequently, the system extracts the wire production process parameters corresponding to the defect area, such as rolling temperature curve, alloy ratio record, cooling rate, and drawing tension change, and correlates and matches the defect severity with the corresponding process data through cluster analysis, thereby identifying the key process variables that cause defects and prioritizing them to provide a basis for process optimization.

[0033] Step S5: Generate wire manufacturing process optimization data based on the wire production process associated data; apply the wire manufacturing process optimization data to adjust wire production process parameters, and verify the wire interface quality to obtain wire interface performance evaluation data.

[0034] The embodiment of the present invention generates a set of optimal wire manufacturing process parameters based on machine learning optimization algorithms (such as genetic algorithms combined with random forest regression) for the identified key process variables, such as rolling temperature range offset, high stress drawing stage, etc., to form a wire manufacturing process optimization data set. This data set covers process adjustment suggestions under various production scenarios, such as reducing the drawing tension by 10%, increasing the intermediate annealing temperature by 30°C, and controlling the metal powder particle size below 10 microns. The optimization data is loaded into the process control system of the production line in real time, and the parameters of the temperature control module, tension controller and mixing unit are automatically adjusted. At the same time, an online interface imaging unit is deployed during the production process to verify the interface quality change trend in real time. By comparing the changes in interface defect distribution and the decrease in defect severity scores before and after optimization, the wire interface performance evaluation data is finally obtained. The evaluation data includes indicators such as defect reduction rate, interface uniformity improvement and overall material strength improvement ratio, verifying the improvement effect of the data analysis method in the actual production line.

[0035] Preferably, step S1 includes the following steps:

[0036] Step S11: Setting up a wire detection device, fixing the composite metal wire to be tested within the detection area of the sensor array, thereby obtaining wire cross-sectional image data and surface feature data;

[0037] In an embodiment of the present invention, a wire detection device is deployed on an experimental detection platform, wherein the wire detection device includes a high-precision fixing fixture, an industrial camera system, and a multi-channel surface sensor array. The composite metal wire to be tested (for example, a 2.5 mm diameter, nickel-titanium alloy copper-clad structure) is fixed to the center of the sensor array of the detection device so that the surface of the wire is at a 90-degree perpendicular relationship to the sensor array. The industrial camera captures the cross-sectional image of the wire at a sampling rate of 60 frames per second, and the light source uses a ring-shaped LED array to reduce shadow interference. At the same time, a surface profile laser scanning module is used to synchronously collect surface feature data of the wire, such as surface waviness, scratch depth, and oxide layer distribution. This step provides high-precision two-dimensional image data and one-dimensional surface property information, providing a multi-source input basis for subsequent morphology reconstruction and interface analysis.

[0038] Step S12: performing morphological feature reconstruction processing on the wire cross-section image data and the surface feature data, thereby obtaining wire interface data of the composite metal;

[0039] The embodiment of the present invention uses a three-dimensional morphology reconstruction algorithm to perform data fusion and reconstruction based on the wire cross-sectional image and surface feature data obtained in step S11. First, the metal material contour and the boundaries of each layer are extracted by image segmentation technology, and then the frequency domain compensation algorithm based on Fourier transform is applied to correct the deformation deviation caused by the curved surface projection. Subsequently, the structured light three-dimensional modeling method is combined with the image overlapping and splicing technology to fuse the data collected from different perspectives to generate a complete three-dimensional structural model. This model not only presents the spatial distribution boundaries of different materials in the composite metal, but also reflects the regional differences in interface bonding strength based on surface roughness, thereby generating wire interface data of the composite metal, including metal interface layer thickness, morphological curvature and interlayer boundary clarity indicators, etc., to provide a basis for subsequent detection parameter selection.

[0040] Step S13: determining the ultrasonic emission position and parameters based on the wire interface data, transmitting a multi-band ultrasonic signal to the fixed composite metal wire, receiving the wire interface echo signal, and calculating the ultrasonic response parameters based on the echo signal intensity, phase, and delay time;

[0041] According to the wire interface data extracted in step S12, the embodiment of the present invention identifies the material interface area and the possible structural discontinuity area in the three-dimensional model, determines that the ultrasonic emission point is located in these critical interface areas, and sets the adaptive multi-band ultrasonic parameters. Specifically, a 5MHz to 15MHz adjustable frequency ultrasonic transducer is used for non-contact excitation, and the emission angle is controlled at 30 degrees with the wire axis to enhance the interface wave coupling effect. During the acquisition process, the intensity (reflecting the interface reflection ability), phase (reflecting the material impedance matching) and delay time (reflecting the interface thickness change) of the echo signal are recorded by the array receiving transducer. Based on the sound wave propagation time and speed calculation formula, the ultrasonic response parameter set corresponding to each emission point is comprehensively extracted to quantify the bonding condition of each interface position and the internal defect clues.

[0042] Step S14: determining an eddy current detection path based on the wire interface data, scanning the wire using a multi-frequency eddy current sensor, measuring changes in the electromagnetic field response at different positions of the wire, and calculating electromagnetic permeability distribution data based on the changes in the electromagnetic field response;

[0043] The embodiment of the present invention continues to use the wire interface data of step S12, selects areas with complex interface structures and obvious material changes as key detection paths, designs eddy current detection paths and deploys multi-frequency eddy current sensors. A sweep frequency excitation method with a frequency range of 10kHz to 1MHz is used to perform a circular scan on the wire, and the detection period of each scanning point is 5 milliseconds. Since different metal layers have different electromagnetic permeabilities, the changes in induced voltage collected by the sensor during the detection process are used to infer their local electromagnetic characteristics. The response curve of the induced electric field with frequency is calculated by Fourier analysis to obtain the electromagnetic conductivity distribution data. The data expresses the spatial changes in the electromagnetic properties of each region in the form of a two-dimensional matrix, reflecting the characteristics of hidden defects such as inclusions, interface debonding and material impurity distribution.

[0044] Step S15: setting X-ray scanning parameters based on the wire interface data, performing a tomographic scan of the wire using low-energy X-rays, and generating an interface density gradient map;

[0045] The embodiment of the present invention refers to the interface model obtained in step S12, sets the X-ray scanning parameters for the wire, and gives priority to low-energy X-rays (energy range of 20keV to 50keV) to enhance the imaging capability of interface density details, while reducing radiation damage to equipment and materials. The wire is placed in a microfocus X-ray tomography scanner, the rotation angle step is set to 1 degree, the scanning resolution is set to 5 microns, and a complete attenuation image sequence of multiple sections of the wire is obtained. A volume rendering method based on the Bayesian reconstruction algorithm is used to perform tomographic reconstruction on the slice data to generate a density gradient map, which shows the changing trend of the density transition zone between different metal layers. The density mutation area usually corresponds to interface voids, unfused or high internal stress accumulation areas. This map further improves the observation perspective of the physical state of the interface.

[0046] Step S16: combining the ultrasonic response parameters, the electromagnetic permeability distribution data, and the interface density gradient map to obtain a wire interface characteristic matrix;

[0047] In an embodiment of the present invention, the ultrasonic response parameters in step S13 (such as the echo intensity spectrum and delay time difference of each interface), the electromagnetic conductivity distribution data in step S14 (reflecting the spatial distribution of material conductivity differences), and the interface density gradient map in step S15 (describing density continuity) are uniformly subjected to feature fusion processing. The fusion method uses feature vector splicing and normalization operations to ensure that various indicators are compared at the same scale to generate a wire interface feature matrix. Each row of the matrix corresponds to a record of multiple physical properties of the composite metal under a certain cross section or scanning path of the wire, including but not limited to local bonding strength indicators, electromagnetic shielding capability evaluation values, structural consistency scores, etc., forming a data vector set with characterization dimensions as high as dozens, which is the basic data for subsequent modeling.

[0048] Step S17: constructing a wire interface feature model according to the wire interface feature matrix.

[0049] The embodiment of the present invention constructs a wire interface feature model based on the wire interface feature matrix generated in step S16, and adopts the self-organizing map (SOM) neural network in machine learning for training. The self-organizing map can map the high-dimensional feature matrix to a two-dimensional topological map in an unsupervised manner, reflecting the similarity and abnormality of the interface structure in different areas. During the training process, each feature vector is used as a network input node, and the network gradually clusters the characteristic patterns of the wire interface in spatial distribution through weight adjustment. The wire interface feature model finally constructed has visualization capabilities and can mark abnormal clustering areas as potential defect focus areas, providing a structured and scalable data expression framework for subsequent defect identification, model correction and performance evaluation.

[0050] It is particularly important that step S15 includes the following steps:

[0051] Step S151: Analyze the material composition information in the wire interface data, determine the optimal X-ray energy range and dose parameters, and generate X-ray scanning configuration data;

[0052] The embodiment of the present invention analyzes the wire interface data collected in the early stage, focusing on extracting information about the different components of the metal composite wire, such as the types of common composite materials such as aluminum, copper, and nickel, and their relative thicknesses. Based on the differences in the penetration ability of X-rays in different metals, the energy spectrum calculation model is used to select the optimal X-ray energy range. For example, for a composite wire with an outer layer of aluminum and an inner core of copper, in order to ensure that the X-rays can penetrate the copper core without over-exposure to the aluminum, an energy range of 80 to 120 kiloelectron volts is selected, and the dose is controlled at 3 milliamperes per second to prevent material heating and image overexposure. At the same time, combined with the maximum outer diameter of the wire (such as 2 mm) and the scanning time requirement (no more than 20 seconds), the scanning speed and exposure time are set, and finally "X-ray scanning configuration data" containing ray energy, dose, voltage, current and time window are generated as the basis for subsequent detection system parameter settings.

[0053] Step S152: adjusting the emission angle and spot size of the X-ray source according to the X-ray scanning configuration data to adapt to the geometric characteristics of the wire, thereby forming X-ray detection parameters;

[0054] The embodiment of the present invention adjusts the geometric emission parameters of the X-ray source in the detection system based on the X-ray scanning configuration data obtained in step S151. Considering that the wire is a slender cylindrical structure, in order to obtain a high-resolution interface image, the emission angle of the X-ray source is set to a 360-degree circular scanning mode perpendicular to the axis of the wire. In order to adapt to the smaller diameter structure of the wire, the X-ray focus size (ie, "spot size") is set to between 0.1 mm and 0.2 mm to improve imaging clarity and interface recognition capabilities. In specific operations, an X-ray source with micro-focus control capability is used, and the scanning path and focus parameters are set in the control system to form X-ray detection parameters including the emission angle range, the step size per angle, the spot size, etc., which are used to guide the subsequent scanning and acquisition process.

[0055] Step S153: performing circular rotation scanning on the composite metal wire based on the X-ray detection parameters to collect X-ray transmission intensity data of the wire at different angles;

[0056] Based on the set X-ray detection parameters, the embodiment of the present invention fixes the metal wire in a rotating platform, rotates it 360 degrees around its own axis, and the X-ray source and detector perform a circular scan on it. One frame of image data is collected for every 1 degree of rotation, and a total of 360 frames of X-ray transmission images are collected. Each frame of the image records the transmission intensity distribution of the wire along the X-ray direction at that angle. A high-sensitivity flat-panel detector with a pixel resolution of 100 pixels per millimeter is used to ensure that the details of the interface structure are captured. Dynamic exposure control is required in this process to ensure that evenly exposed images are obtained at locations of different thicknesses. The final X-ray transmission intensity data is represented in the form of a two-dimensional grayscale image. The grayscale value in each frame of the image corresponds to the degree of attenuation of the ray at that position, thereby indirectly reflecting the material density and interface conditions at that location.

[0057] Step S154: performing denoising and contrast enhancement on the X-ray transmission intensity data to obtain original density image data of the internal structure of the wire rod;

[0058] The embodiment of the present invention performs image denoising and enhancement processing on the X-ray transmission intensity image data collected in step S153. First, a median filtering algorithm is used to remove high-frequency noise, and in particular, a filtering process is performed on the edge area of the image to retain edge characteristics; then, a contrast-limited adaptive histogram equalization method (CLAHE) is applied to enhance the grayscale contrast of the interface area, so that the interface density mutation area is more clearly distinguishable in the image. In this process, image artifact suppression processing is also performed to eliminate the stripe interference caused by mechanical offset during the rotational scanning process. The processed image set is the "original density image data", which can be used as input for subsequent tomographic reconstruction, retaining information on the micro-structural changes inside the wire.

[0059] Step S155: performing tomographic reconstruction calculation based on the original density image data to generate a density distribution tomogram of the wire cross section;

[0060] The embodiment of the present invention uses the raw density image data obtained in step S154 to perform tomographic reconstruction calculations. Specifically, a back-projection algorithm or filtered back-projection (FBP) method is used to restore the two-dimensional density distribution image of each cross-section of the wire from the multi-angle projection image. Specifically, the pixels at the same scanning angle in each frame of the image are integrated back-projected along a set axis, and a bandpass filtering operation is simultaneously performed to suppress low-frequency drift. Several cross-sectional tomograms are reconstructed. The image size is set to 512×512 pixels, and the spatial resolution reaches 20 microns / pixel. In a test of a composite aluminum-copper wire, a total of 200 cross-sectional density distribution tomograms were generated, which effectively reflect the transition zone structure between the central copper core and the outer aluminum cladding of the wire.

[0061] Step S156: extracting the metal interface region from the density distribution tomogram, analyzing the density variation characteristics of the materials on both sides of the interface, and constructing interface density transition zone characteristic data;

[0062] The embodiment of the present invention extracts interface area information from the tomographic reconstruction image, mainly identifying the geometric position of the material interface and its transition characteristics. Through the image gradient analysis method, the grayscale mutation area is identified in the tomogram, which is the junction of the two metal materials. Subsequently, the grayscale change curve within a certain width (such as ±50 microns) is extracted in the left and right areas adjacent to the interface, and its density change trend is fitted to construct "interface density transition zone characteristic data". This data reflects the density smoothness of the material when it transitions from one metal to another, and is an important basis for judging the interface fusion quality and welding integrity. In specific applications, when the width of the interface density transition zone exceeds 100 microns, there may be problems of insufficient welding or interface embrittlement.

[0063] Step S157: Calculate the interface region density change rate based on the interface density transition zone characteristic data to generate an interface density gradient map.

[0064] The embodiment of the present invention is based on the characteristic data of the interface density transition zone and uses the numerical differentiation method to calculate the rate of change of density with position, that is, the density gradient. The density change rate of each point is calculated by the central difference or third-order finite difference method, and then the "interface density gradient map" of the entire interface area is constructed. The map is presented in the form of a two-dimensional heat map, and the high gradient area usually represents the location where the material interface is suddenly changed or a defect exists. In a specific example, if the density gradient of a certain area exceeds the set threshold (such as the rate of change per millimeter is greater than 500 kilograms per cubic meter), it is marked as a potential abnormal point, and other physical detection methods can be used to locate and confirm it later. The interface density gradient map not only provides a quantitative basis for subsequent interface characteristic analysis, but is also used in process control to evaluate welding process stability and material compatibility.

[0065] Preferably, step S17 includes the following steps:

[0066] Step S171: performing data standardization processing on the wire interface characteristic matrix, incorporating different physical quantities into a unified measurement system, and generating standardized interface characteristic data;

[0067] After obtaining the "interface density gradient map" generated in step S157, the present embodiment first extracts key feature data for subsequent analysis and modeling to form a "wire interface feature matrix." This matrix typically includes multiple physical dimension dimensions, such as the width of the interface density transition zone, the maximum density gradient value, the mean density gradient, the transition region symmetry index, the interface position offset in the tomogram, and the interface roughness factor. Each row represents a wire sample, and each column represents a feature dimension. Because the physical units of these feature dimensions vary significantly—for example, density is measured in kilograms per cubic meter, position is measured in micrometers, and gradient is the ratio of density to length—directly using these features for modeling and analysis would result in inconsistent dimensions and imbalanced dimensionality. To address this issue, the feature matrix requires data normalization. Z-score normalization is employed: for each column of feature data, the sample mean is subtracted and divided by the standard deviation, resulting in a zero mean and unit standard deviation for each feature dimension after normalization. This process is implemented using the Pandas library in Python combined with the StandardScaler module in Scikit-learn. Taking a batch of composite aluminum-copper wire as an example, the original range of the interface roughness factor is between 0.2 and 3.8, while the density gradient value is between 500 and 3000. Without standardization, the model will tend to interpret features with a large range of variation, resulting in weight shift. After standardization, the physical characteristics of these different units are unified into dimensionless standard values, thus generating "standardized interface feature data", which provides a reliable and balanced data input basis for subsequent interface quality assessment models or cluster analysis.

[0068] Step S172: establishing a three-dimensional coordinate system based on the standardized interface feature data, mapping the wire geometry information into the spatial coordinate system, and forming wire interface spatial mapping data;

[0069] The embodiment of the present invention performs a standardization process on the wire interface feature matrix obtained in step S16, that is, unifies the numerical range of various data in the ultrasonic response parameters, electromagnetic permeability distribution data, and interface density gradient map, so that data of different physical quantity dimensions can be spatially mapped under the same analysis scale. Subsequently, a three-dimensional coordinate system is established with the wire center axis as the Z axis and the cross-sectional direction as the XY axis, and each type of standardized interface feature data is mapped to the three-dimensional coordinate system according to its spatial position, forming "wire interface space mapping data" with spatial positioning attributes. This data can be represented in the form of a sparse tensor, and each tensor unit records multiple interface feature values at the spatial point. In a specific application scenario, for a composite metal wire with a diameter of 1.2 mm and a length of 200 mm, the spatial coordinate system is divided into 200 equally spaced cross sections, each cross section is divided into 36 angular directions and 10 radial layers, and a total of 72,000 spatial mapping points are obtained.

[0070] Step S173: constructing a basic topological structure of the interface using the wire interface space mapping data, determining the continuity and morphological characteristics of the interface, and generating interface topological structure data;

[0071] The embodiment of the present invention utilizes the above-constructed spatial mapping data, adopts regional connectivity analysis and isosurface reconstruction method to extract continuous interface feature areas, identifies boundary surfaces of different interface morphologies through three-dimensional boundary tracking algorithm, and establishes the basic topological structure of the interface. The so-called "basic interface topological structure" refers to the construction of a graph structure model reflecting the overall morphology of the interface based on the connectivity, adjacency and curvature information of the interface features of different materials of the wire in space. In this process, the characteristic gradient change between each spatial point and its adjacent points is calculated to determine whether they belong to the same topologically connected area, and then divide different interface topological blocks. In practical applications, if the electromagnetic permeability gradient exceeds the set threshold of 0.2 Siemens / mm, it is regarded as the boundary point of different interfaces. The interface topological structure data constructed in this way can effectively reflect the continuity, nestedness and local abnormal deformation of the material interface.

[0072] Step S174: analyzing the spatial distribution pattern of interface features based on the interface topology data, identifying the feature change pattern of the interface region, and forming interface region feature clustering data;

[0073] The embodiment of the present invention performs spatial statistical modeling based on interface topological structure data, and uses a density clustering algorithm (such as density-based spatial clustering DBSCAN) to identify regional groups with similar interface feature change trends in three-dimensional space, namely "interface region feature clustering data". In the clustering process, the interface feature vector of each point is used as the clustering input, and the Euclidean distance between points and the feature gradient change threshold are combined to determine whether they belong to the same class. In a specific implementation, areas where the interface density changes continuously by less than 5% in the thickness direction and there is a significant density mutation (more than 15%) in the transverse direction are clustered into one category to identify potential delamination, cracks, and dissimilar metal contact defect areas. The final output is a structured data set containing the cluster center position, regional shape description, and corresponding interface type label.

[0074] Step S175: Screening key characteristic parameters sensitive to interface quality based on the interface region feature clustering data to obtain a set of interface key characteristic parameters;

[0075] The embodiment of the present invention further performs a sensitivity analysis on the clustering results of the identified interface regions, and selects characteristic dimensions that have a significant impact on changes in quality indicators such as interface strength, interface integrity or interface corrosion resistance as key characteristic parameters. For example, the principal component regression (PCR) method is used to analyze the correlation between each interface feature and the experimental results of the interface strength of known samples, and the ultrasonic phase delay, electromagnetic conductivity gradient and density mutation rate are determined to be the main influencing factors. The "interface key characteristic parameter set" finally obtained is a core variable set for interface quality modeling, which in practice includes parameter items such as "maximum interface conductivity mutation position", "maximum interface thickness change rate section", "minimum interface density gradient length", etc., with a total of about 15 dimensions.

[0076] Step S176: constructing a feature vector space using the interface key feature parameter set, and performing dimensionality reduction processing to generate interface feature dimensionality reduction data;

[0077] The embodiment of the present invention uses the above-mentioned interface key feature parameter set to construct a feature vector space, in which each sample corresponds to a vector composed of 15 feature parameters. In order to improve the efficiency of the model and reduce dimensional redundancy, linear discriminant analysis (LDA) is used for dimensionality reduction processing, taking into account both the discrimination ability and information integrity, to generate 3-5 dimensional interface feature dimensionality reduction data. In the specific operation, the feature parameters are subjected to covariance matrix analysis, the main axis direction is extracted, and the top few main axes are selected according to the standard of information retention rate not less than 95%. The feature data after dimensionality reduction is convenient for subsequent use in mathematical modeling and physical simulation, thereby improving analysis efficiency and accuracy.

[0078] Step S177: establishing a wire interface mathematical model based on the interface feature dimensionality reduction data, and performing physical property constraints on the wire interface mathematical model to obtain a wire interface feature model.

[0079] The embodiment of the present invention constructs a mathematical model for describing the spatial characteristics and physical properties of the wire interface based on the interface feature dimensionality reduction data, namely the "wire interface mathematical model". The model can be in the form of a combination of multivariable functions, describing the functional relationship between the key parameters of the interface in the spatial distribution and the interface stress field and bonding strength. Then, physical constraints are imposed on the model according to the physical properties of the metal, such as interface continuity constraints (that is, the interface strength function is continuous within the topological structure) and material mechanics constraints (combined with Hooke's law and the interface bonding model) to ensure that the constructed model has actual physical meaning. Finally, the "wire interface feature model" is obtained, which can be used in application scenarios such as interface performance prediction, production process control and defect location. In the quality assessment of a certain aluminum-copper composite wire, the model predicted the interface debonding area and the actual ultrasonic detection results with a consistency rate of more than 93%, which has strong practicality and accuracy.

[0080] Preferably, step S2 includes the following steps:

[0081] Step S21: establishing a reference area of the wire interface feature model, selecting an interface section without obvious defects as a reference standard, and generating wire interface reference feature data;

[0082] After completing the construction of the interface feature model in the embodiment of the present invention (as shown in step S177), it is first necessary to define a representative benchmark reference area. This area is usually selected from a sample segment in a large batch of wires that has been manually or automatically detected and confirmed to have no obvious defects, uniform interface structure, smooth density gradient, and good surface continuity. In the specific implementation, a sample screening method can be adopted, combining the previous standardized feature data with the original tomogram, taking the feature stability value interval as the screening condition (such as density gradient between 800 and 1200, interface roughness factor less than 1.0), and combining the three-dimensional image to judge the interface continuity. Finally, 5 to 10 length segments are selected from each wire for manual annotation and confirmation, and their interface feature model parameters are extracted after merging, and the wire interface benchmark feature data is formed after averaging. This data will serve as a reference template for all subsequent test samples.

[0083] Step S22: performing point-to-point comparison between each detection area in the wire interface feature model and the wire interface reference feature data, calculating feature difference values, and forming original wire interface differential data;

[0084] In step S22 of the embodiment of the present invention, the reference feature data generated in S21 is used as a reference, and all detection areas in the wire interface feature model are traversed in sequence, and a point-to-point feature comparison mechanism is constructed for each area. This mechanism refers to comparing each dimension of the standard feature vector and the feature vector of the area to be measured at the same position or at equal intervals, and calculating the feature difference value. The difference value can be calculated using the absolute difference or relative difference method, and the specific selection depends on the purpose of the detection. Taking aluminum-copper composite wire as an example, if the detection target is to identify slight delamination or interface discontinuity, the relative difference is more sensitive. Ultimately, all detection sections generate a set of wire interface original differential data, and the data structure is a one-dimensional or two-dimensional matrix, and each item represents the degree of difference between a certain detection position and the reference standard.

[0085] Step S23: performing statistical normalization processing on the original differential data of the wire interface and performing neighborhood enhancement processing to obtain enhanced differential data of the wire interface;

[0086] In order to eliminate misjudgments caused by deviations in individual regions, the embodiments of the present invention need to perform statistical normalization processing on the original differential data of the wire interface. This processing standardizes the original difference values according to the overall distribution of the region and unifies them to the range of 0 to 1 or -1 to 1. A common method is to subtract the mean from the differential data and divide it by the standard deviation or maximum value to ensure that the difference values between different batches are comparable. After normalization, in order to highlight the spatial continuity and regional aggregation effect of the difference signal, it is also necessary to implement neighborhood enhancement processing, that is, to use a sliding window to calculate the difference mean or median of the local area in the differential map, thereby enhancing the regional signal with spatial consistency characteristics. The final output of the wire interface enhanced differential data is a spatially enhanced expression of the original difference, which can more accurately reflect the significance of the potential abnormal area of the interface.

[0087] Step S24: setting multi-level threshold conditions based on the wire interface enhanced differential data, and performing hierarchical screening on the differential signals to form wire interface differential signal conversion data;

[0088] The embodiment of the present invention performs graded processing for the strength of the enhanced differential data. The specific method is to set multi-level threshold conditions, for example, to set three levels: low difference (less than 0.3), medium difference (0.3-0.6), and high difference (greater than 0.6), and to mark the spatial distribution position of each level of difference to form a graded signal diagram. In practical applications, the setting of multi-level thresholds is based on a large number of sample statistical analysis and experimental verification in the early stage to ensure the ability to detect both minor anomalies and serious defects. Subsequently, the differential signal conversion is performed on the difference area of each level, that is, the difference signal is converted into a labeled result, indicating the abnormality level of the position, thereby obtaining the wire interface differential signal conversion data, which provides a discretization basis for subsequent pattern recognition and defect determination.

[0089] Step S25: Identifying region groups with similar change patterns based on the wire interface differential signal conversion data, thereby extracting wire interface change characteristics;

[0090] The embodiment of the present invention utilizes differential signal conversion data and applies cluster analysis methods such as K-means or DBSCAN to extract groups of regions with similar change patterns. The core of this step is to find a combination of regions that are continuous in space or similar in feature change trends to determine whether there are systematic interface anomalies. The specific process includes: extracting the spatial position, difference intensity and morphological characteristics (such as length, area, directionality, etc.) of each difference region, and performing multi-dimensional clustering to eventually form several interface change pattern groups. These groups are regarded as candidate areas where there may be defect evolution trends in the wire interface, and their statistical characteristics will be used to further evaluate the credibility and risk level of the defects.

[0091] Step S26: Perform spatial mapping based on the wire interface variation characteristics, determine the boundaries and ranges of high-confidence defect areas, and construct a preliminary wire interface defect indication map.

[0092] In an embodiment of the present invention, the interface change groups identified in step S25 are mapped in three-dimensional space, that is, the two-dimensional differential analysis results are restored to the original interface feature model or three-dimensional tomogram, and the spatial boundary, volume, and relative position of each change group to the reference area are calculated to comprehensively determine whether it belongs to a high-confidence defect area. If a group has a high degree of difference, strong spatial ductility, and a structural morphology that is consistent with a typical defect (such as a strip-like extension along the interface direction), its boundary is extracted to form a preliminary indicator map of wire interface defects. This map is an important intermediate result of wire interface quality assessment and can be used in multiple application scenarios such as subsequent manual review, automatic defect annotation, or machine learning model training.

[0093] Preferably, the multi-scale decomposition processing based on the preliminary indication map of wire interface defects in step S3 includes:

[0094] Perform multi-resolution segmentation on the preliminary indication image of wire interface defects, divide the image into macro-scale, meso-scale and micro-scale detection areas, and generate multi-level defect decomposition data;

[0095] Perform morphological expansion processing on the macro-scale detection area to enhance the continuity characteristics of the interface crack and generate macro-defect connected domain data;

[0096] Adaptive threshold segmentation is performed on the mesoscopic detection area to separate the grain boundary slip band and the second phase precipitation characteristics, generating mesoscopic defect separation data;

[0097] Perform Laplace sharpening filtering on the micro-scale inspection area to generate micro-defect enhancement data;

[0098] Input multi-level defect decomposition data, macroscopic defect connected domain data, mesoscopic defect separation data, and microscopic defect enhancement data into a multi-channel data fusion network to generate cross-scale defect correlation data;

[0099] Construct a three-dimensional defect space map based on cross-scale defect correlation data, mark the distribution density of defects in the X / Y / Z axes, and generate a defect space distribution vector;

[0100] Perform principal component analysis on the defect spatial distribution vector to generate a defect core feature data set;

[0101] The defect core feature dataset is matched with the preset defect type database by cosine similarity, and feature combinations with similarity > 85% are screened to generate interface defect feature data.

[0102] In this embodiment, the preliminary wire interface defect indication map obtained in step S26 is used as input. This map contains the three-dimensional coordinates and image intensity values of the interface defect region. To better identify defect features at different scales, a strategy based on image pyramid and multi-scale structural analysis is adopted. The entire image is decomposed into three levels of spatial resolution: a macroscale resolution of 1 mm is used to observe overall structural discontinuities and interface deformation trends; a mesoscale resolution of 200 microns is used to identify grain boundary distortion or small-sized precipitates in the microstructure; and a microscale resolution of 50 microns or less is used to capture microcracks, microvoids, or areas of atomic dislocation concentration. The segmentation method is based on a sliding window block with spatial scale constraints. Each sub-image is generated for each level of spatial resolution, with each sub-image labeled for its scale. The final multi-level defect decomposition data output includes the image block index, spatial coordinate range, scale label, and preliminary defect intensity distribution information at each scale, which is used in the subsequent multi-scale feature extraction process. The macroscale detection area image block set generated in the previous step is selected, and the initial crack contours within these areas are subjected to morphological dilation processing to improve their structural coherence. Morphological dilation is a structural enhancement technique that improves the continuity and connectivity of fracture paths by expanding pixels around crack edges. In this example, a 5×5 structuring element was selected as the dilation kernel. This elliptical structuring element adapts to the directional changes in the extension of interface cracks. During the operation, each macroscopic image block is binarized to extract the defect boundary, and then a dilation operation is performed to determine whether the connected domain formed after dilation spans the main axial direction of the interface. If so, its start and end coordinates, connectivity length, and connectivity strength are recorded to generate a macroscopic defect connected domain dataset. This dataset can be used to subsequently determine whether the crack has formed a macroscopically visible defect channel and is of great reference value in fatigue life prediction of metal conductors. The mesoscopic image blocks divided in the previous step are processed. Each image covers an area of approximately 200 microns and is primarily used to detect material microstructural anomalies. For grain boundary slip bands (often manifested as areas with directional changes in grayscale intensity) and second-phase precipitates (often manifested as local bright or dark spots) at this scale, this example uses an adaptive threshold segmentation method based on local grayscale statistics. Specifically, a sliding window process is performed on each image block (window size is 20×20 pixels), and the local mean and local variance are calculated respectively. The local mean plus or minus twice the standard deviation is used as the adaptive threshold standard to dynamically determine the edge area. The image is then divided into background, slip band area, and precipitate area according to the threshold. The output mesoscopic defect separation data is saved in the form of a label map, indicating the location, size, directionality, and other information of the three types of areas. This data is particularly suitable for evaluating the structural stability and interface composition compatibility of the heat-affected zone during interface processing. For the image blocks of the microscale detection area, the resolution of each image is 50 microns or less.Because defects at the microscale often manifest as low-contrast microcracks or lattice perturbations, traditional enhancement methods struggle to clearly extract edge features. Therefore, a Laplacian sharpening filter algorithm is employed. This algorithm, based on the edge enhancement mechanism of second-order derivatives, enhances edge responses by detecting locations with high rates of change in the grayscale image. A 3×3 Laplacian convolution kernel is used in this operation, combined with the original grayscale image for filtering. The resulting image is then weighted and fused with the original image (weights set to 0.7 for the original image and 0.3 for the sharpened image) to enhance defect edges while preserving the overall image structure. The micro-defect enhancement data, which records the spatial position index of the enhanced image, enhancement weight parameters, and a mask of significant edge regions, is used to subsequently identify the boundary morphology of submicron defects, making it particularly useful in electron beam inspection and X-ray tomography. A multi-channel convolutional fusion network (MCFN) is employed to unify the defect feature information at these scales. The network structure contains four input channels, corresponding to macroscopic image patches, mesoscopic image labels, microscopic enhanced images, and scale decomposition metadata. After each channel extracts features through an independent convolutional layer, the features are spliced in the intermediate fusion layer. Then, a two-layer attention mechanism module is used to identify the cross-relationship between scales, and finally the fused cross-scale feature map is output. During the network training phase, manually annotated defect maps are used as supervisory signals. The training loss function incorporates cross-entropy and structural similarity error terms to optimize edge fidelity and structural consistency at different scales. The cross-scale defect association data records the spatial distribution map, eigenvalue matrix, and significance score of each defect area at multiple scales, and is the core input for subsequent three-dimensional mapping and type recognition. The defect association data output by the fusion network is matched with the initial three-dimensional coordinate system, and each defect area is remapped to the original wire voxel grid to construct a defect point cloud map with spatial location attributes. By counting the spatial coverage and density changes of each defect in the X, Y, and Z directions (the unit is the ratio of the number of defect points to the spatial volume), density vectors in the three axial directions are generated respectively. The length of each vector is equal to the number of spatial units divided in the corresponding direction, and the vector value is the density of defect points within the unit. Finally, the defect spatial distribution vector is integrated to form its data format as a triple array, which records the starting and ending positions, peak density and standard deviation of the high-density area in each direction. In practical applications, it can be used to determine spatial evolution information such as defect aggregation direction and crack propagation tendency. Based on the statistical principal component analysis (PCA) method, the three-dimensional defect distribution vector obtained in the previous step is subjected to dimensionality reduction processing.The process first concatenates density vectors in the X, Y, and Z directions into a long vector set, which is then centered and normalized. The first three principal component directions are then extracted through covariance matrix calculation, with information representing more than 90% of the explained variance retained in the core feature data. The resulting defect core feature dataset contains the principal component vector coefficients, the projections of the original vectors in principal component space, and the explanatory power of each principal component. In practical analysis, this dataset can be used to extract the most representative defect structural features and serves as a key input for constructing defect fingerprints or training defect recognition models. A previously constructed defect type database is introduced, encompassing over ten common defect types, including interface delamination, metallurgical inclusions, thermal stress cracks, and grain boundary mismatches. Each defect type is represented by its representative principal component signature. Cosine similarity is used as a measure of similarity during the matching process. The cosine of the angle between the current defect core feature vector and each defect vector in the database is calculated. Types with a similarity greater than 85% are selected as preliminary matching results, and their corresponding labels, similarity scores, and principal component contributions are recorded. The interface defect feature data finally generated includes not only category labels, but also defect intensity levels, feature stability evaluations, and matching confidence levels, which have important application value in scenarios such as quality traceability, failure analysis, and process optimization.

[0103] Preferably, identifying the wire interface defect type characteristics according to the interface defect characteristic data in step S3 includes:

[0104] Extract the defect length, width, depth and area ratio characteristic values from the interface defect feature data to generate the defect basic geometric feature set;

[0105] Compare the defect basic geometric feature set with the standard defect geometric template in the defect type database one by one to calculate the geometric similarity;

[0106] Based on the geometric similarity, a candidate set of defect types with a matching degree of >75% is selected;

[0107] Based on the defect type candidate set, the density distribution of defects in the material interface area, the spacing between adjacent defects, and the arrangement direction are calculated to generate the defect spatial distribution feature vector;

[0108] Defect groups are divided based on density peak detection according to the defect spatial distribution feature vector to generate defect spatial clustering data;

[0109] Extract defect edge morphological features based on defect spatial clustering data, quantify edge roughness, fractal dimension and curvature change rate, and generate defect edge feature parameter set;

[0110] Perform secondary matching based on the defect edge feature parameter set and the defect type database to correct misjudgments in the preliminary defect classification data and generate optimized defect classification results;

[0111] The optimized defect classification results are judged according to the preset defect type judgment rules to generate wire interface defect type features including defect type code, confidence level and position coordinates.

[0112] The embodiment of the present invention uses an image measurement algorithm to quantify the geometric projection contour of the defect in the image. Specifically, a contour analysis method based on boundary extraction is used to extract the maximum extension direction length of the defect as the defect length and the maximum width in the perpendicular direction as the defect width. The defect extension information in the Z-axis direction in the three-dimensional defect space mapping data is further combined to extract the defect depth; at the same time, the projection area of the defect in the entire wire interface image is ratioed to the total area of the interface region to obtain the defect area ratio. Based on this, a basic geometric feature set of defects is constructed, covering length (unit micron), width (unit micron), depth (unit micron) and area ratio (percentage) for subsequent geometric template matching analysis. Based on the generated basic geometric feature set of defects, the embodiment introduces the preset defect geometric templates in the standard defect type database, such as three-dimensional geometric structure templates such as typical grain boundary cracks, second phase cluster precipitation, and slip band cracking, and uses a matching algorithm combining Hausdorff distance and shape context descriptor to calculate the geometric morphology similarity of each defect sample. The specific operation includes first standardizing the size of the defect sample and the template, then extracting their boundary shape distribution and angle information respectively, and fitting them through the shape distribution function to obtain the similarity score between each pair of defect samples and the template. Screen the templates with a similarity of more than 75% as a candidate set of defect types for preliminary matching. After screening out the candidate set with a geometric morphology matching degree greater than 75%, the embodiment calculates the distribution density of each type of defect per unit area of the image based on the spatial position coordinate information of the defect in the material interface area, and at the same time performs Euclidean measurement on the center distance between the defects to obtain the spacing between adjacent defects, and then calculates the overall arrangement direction angle of the defects through the main direction vector fitting algorithm. This method uses local principal component analysis (Local PCA) to solve the direction vector of multiple defect point clouds to form a set of feature vectors reflecting the spatial structural distribution of defects, including density value (unit number / square millimeter), average spacing (unit micrometer) and main direction (unit degree) as a structural representation of the spatial distribution of defects. Subsequently, in the embodiment, a cluster analysis is performed on the aforementioned defect space distribution feature vectors, and a method based on density peak detection (Density Peaks Clustering, DPC) is used to identify density centers and outliers to form defect group division results. This method calculates the density value of each defect point and the minimum distance to the sample with higher density, thereby identifying the cluster center and realizing the natural clustering division of defects in the interface area. For example, in a 200×200 micron wire interface image, there are about 30 spatially continuous defect points. The algorithm can identify it as a high-density cluster area and generate defect space clustering data containing multiple cluster category labels.Based on the generated defect space clustering data, the embodiment further extracts the edge morphological features of each type of defect group, calculates the edge roughness (evaluates the contour smoothness by the boundary fluctuation coefficient), the fractal dimension (measures the boundary complexity by the box dimension method) and the curvature change rate (reflects the degree of sudden change in shape based on the standard deviation of the continuous change of curvature), and quantifies the edge structure complexity of each defect area. These features help to distinguish different defect types such as grain boundary cracks (edge rules) and micro corrosion points (edge roughness, curvature mutation), forming a defect edge feature parameter set for further classification optimization. After obtaining the defect edge feature parameter set, the embodiment performs a secondary comparison with the edge morphology template in the aforementioned standard defect type database, and uses the support vector machine (SVM) classifier combined with the K nearest neighbor algorithm to correct the misjudgment of low-confidence samples in the preliminary classification results, especially for defects with similar geometry but large differences in edge features, such as slip band cracking and grain corrosion, which are similar in length but have significant differences in fractal dimension. After comparison, the defect classification results are recalibrated to generate an optimized defect classification result data set. In the embodiment, based on the preset defect type judgment rules in the metal wire defect recognition system (for example, crack defects are required to have a length greater than 50 microns and a consistent distribution direction, and corrosion defects account for less than 3% of the area but have large edge roughness), the optimized defect classification results are subjected to rule judgment, and the samples that meet the rule matching are marked as specific defect types. The output includes structured result data including the defect type code (such as CRK representing crack), recognition confidence (such as 0.93) and position coordinates (such as x=112.4, y=85.2), forming wire interface defect type characteristics that can be used for subsequent quality control and defect prediction modeling.

[0113] It is particularly important that the defect type determination rules are as follows:

[0114] If the defect aspect ratio is greater than 5, the arrangement direction angle ∈ [60°, 120°], and the cluster label is a continuous group, it is marked as an interface through-crack;

[0115] If the defect area ratio is >3%, the edge roughness is >2.5μm and the cluster density is <3 / mm 2 , it is marked as local material peeling;

[0116] If the defect spacing is less than 0.1 mm, the fractal dimension is greater than 1.6, and the cluster density is greater than 5 / mm 2 , it is marked as diffuse pore aggregation.

[0117] The embodiment of the present invention calculates the geometric shape characteristics of the defect based on the aspect ratio of the defect, and this value is obtained by measuring the ratio of the longest and shortest axes of the defect. In this example, if the value is greater than 5, and the arrangement direction angle of the defect is between 60° and 120°, and the cluster label of the defect is a continuous group, it means that the morphology of the defect is consistent with the interface penetration crack. Interface penetration cracks usually appear as cracks with a long length and strong directional consistency. In order to further determine the arrangement direction of the defect, the embodiment uses a rotation-invariant feature descriptor (such as Hough transform) to calculate the main direction of the defect and determine whether it falls within the specified direction interval. Based on this information, combined with the group connectivity identified in the cluster analysis, the defect can be accurately marked as an interface penetration crack, which is usually a typical defect in the metal wire processing process. For the identification of local material peeling, the embodiment first calculates the area ratio of the defect, that is, the ratio of the defect area to the entire image. If the area ratio is greater than 3%, and the edge roughness of the defect exceeds 2.5μm, and the cluster density is less than 3 / mm 2 , then the defect is determined to be local material peeling. The edge roughness is obtained by contour analysis of the defect edge. The specific calculation method adopts an algorithm based on edge fluctuation analysis, and uses wavelet transform to decompose the defect contour to obtain the roughness feature. The cluster density is obtained by calculating the number of defects per unit area in the defect group. In this scenario, local material peeling is often manifested as peeling or corrosion of the material surface, so the edge complexity is high, but the density is low, and the peeling area is usually large, which meets the above standards. Through the above steps, the embodiment can accurately identify these defect types and mark them. For the identification of diffuse pore aggregation, the embodiment first calculates the minimum distance between defects, which can be obtained by calculating the Euclidean distance of the center points of all defects. If the defect spacing is less than 0.1mm, and the fractal dimension is greater than 1.6, and the cluster density is greater than 5 / mm 2 , then the defect can be determined to be a diffuse pore aggregation. The fractal dimension is obtained by calculating the complexity of the defect outline using the box dimension method, in which the details of the defect boundary and the complexity of the distribution are quantified as a numerical value. Diffuse pore aggregation is usually manifested as a large number of small pores tightly arranged in a local area. These pores usually have a high aggregation density, so during the detection process, the analysis of density and spacing is a key indicator. In this example, combining the characteristics of spacing and fractal dimension can effectively distinguish these small pore clusters and accurately classify them.

[0118] Preferably, the matching and correction of the wire interface defect type characteristics with a preset physical constraint model in step S3 includes:

[0119] Establish a physical constraint model library for wire interface defects, including the physical property parameters and constraints of delamination, pores and inclusions;

[0120] Parametrically compare the wire interface defect type characteristics with the wire interface defect physical constraint model library to generate preliminary defect type matching data;

[0121] Based on the preliminary matching data of the defect type, the physical rationality of the defect characteristics is evaluated, the constraint conditions are verified, and the physical rationality assessment of the defect is formed;

[0122] Perform feature correction based on physical constraint rules on the physical rationality evaluation of defects to obtain defect feature correction data;

[0123] Calculate the quantitative defect characterization parameters including defect size, depth, morphology and distribution characteristics based on the defect feature correction data to generate a set of defect quantitative characterization parameters;

[0124] The defect quantitative characterization parameter set is matched with the defect evolution law in the wire interface defect physical constraint model library, and the defect formation mechanism and development trend are evaluated to generate accurate defect characterization data.

[0125] This embodiment of the present invention establishes a physical constraint model library for wire interface defects, which contains the physical property parameters and associated constraints for various metal wire interface defect types (such as debonding, pores, and inclusions). Specifically, the physical properties of debonding defects may involve their bond strength, area, depth, and mechanical behavior with the base material; pore defects involve porosity, pore size distribution, and their impact on the material's mechanical properties; and inclusion defects include the inclusion's morphology, size, distribution, and its impact on material strength and corrosion resistance. These defect characteristics are concretized into physical constraint models to aid in the evaluation and correction of defect characteristics in subsequent steps. This model library not only includes conventional defect morphological characteristics but also covers relevant physical constraints, such as the material's elastic modulus, plastic deformation behavior, and shear strength of the material interface. Next, the embodiment generates preliminary defect type matching data by parametrically comparing the wire interface defect type characteristics with the wire interface defect physical constraint model library. This process first extracts the physical characteristics of the defect, including its size, morphology, distribution density, and edge roughness. These characteristics are then matched with the parameters defined in the defect physical constraint model library. For example, for debonding defects, matching criteria may include the ratio of the defect area, depth, and matrix interface strength. For porosity defects, the focus is on the relationship between porosity and the material's compressive performance. This comparison generates preliminary defect type matching data, identifying the defects that best match the known defect types. After obtaining preliminary matching data, embodiments further perform a physical plausibility assessment of the defect characteristics. Based on this physical plausibility assessment, each defect's physical properties are checked to see if they meet the corresponding physical constraints. For example, if the depth and width of a defect exceed the limits of the material's mechanical properties, or if the morphology of certain defects does not conform to the assumptions in a known physical constraint model, these defects will be marked as unreasonable. This step primarily serves to screen out defect characteristics that do not conform to physical laws, ensuring the accuracy of subsequent data analysis. Next, based on the results of the physical plausibility assessment, embodiments perform corrections to the defect characteristics based on physical constraint rules. This step corrects the preliminary matching defect characteristics by applying physical constraint rules. For example, if a defect's morphology appears unreasonable in the physical constraint model, the data can be corrected by adjusting the defect's edge shape or size to conform to the physical constraint rules. The corrected defect data is more consistent with actual material behavior, enhancing the accuracy and reliability of subsequent defect analysis. After obtaining the corrected data of the defect characteristics, the embodiment further calculates the quantitative characterization parameters of the defect. These parameters include the size of the defect (such as the length, width, depth, etc. of the defect), the morphology (such as the aspect ratio, edge smoothness, etc.), and the distribution characteristics (such as the uniformity or aggregation of the defect distribution on the material surface). These quantitative parameters can be extracted through morphological analysis, geometric parameter calculation and statistical analysis methods.In specific applications, different image analysis techniques can be combined, such as edge detection-based algorithms to extract the geometric morphology of defects, and cluster analysis-based algorithms to evaluate the distribution characteristics of defects, and finally obtain a quantitative characterization parameter set containing the geometric morphology and distribution of defects. Finally, the embodiment is based on the defect quantitative characterization parameter set and matches it with the defect evolution law in the wire interface defect physical constraint model library. This step mainly analyzes the formation mechanism of defects in metal wires and possible development trends by modeling and simulating the evolution process of defects. For example, peeling defects may gradually expand over time, while pore defects may fuse under certain conditions to form larger pores. By matching the evolution law of defects and the physical constraint model, the formation mechanism of defects can be evaluated and their possible expansion or development trends can be predicted. Ultimately, the generated defect precise characterization data can provide strong support for wire quality control, defect repair and optimization.

[0126] Preferably, step S4 includes the following steps:

[0127] Step S41: spatially mapping the defect precise characterization data based on the axial and radial coordinates of the wire rod to generate wire rod defect spatial positioning data containing three-dimensional defect distribution information;

[0128] The embodiment of the present invention needs to convert the defect feature data on the metal wire interface into a coordinate system suitable for three-dimensional spatial positioning. The specific operation includes mapping the lateral position, longitudinal position and depth value of each defect (such as the depth, width, length, etc. of the defect) into the three-dimensional coordinate system of the metal wire. By using three-dimensional space mapping technology, the spatial position of each defect on the surface of the material and inside it can be obtained. In order to achieve this process, it is possible to use equipment such as laser scanners, CT imaging or X-ray imaging to obtain accurate three-dimensional geometric data of the metal wire, and accurately locate the defect data according to its position and size to generate defect space positioning data containing three-dimensional coordinate information. This data will provide a basis for subsequent defect distribution analysis, quality assessment and process correlation.

[0129] Step S42: Based on the wire defect spatial positioning data, different types of defects are displayed in layers and density statistics are performed to construct a wire interface defect distribution map reflecting the distribution rules of various defects;

[0130] The embodiment of the present invention is based on the wire defect spatial positioning data obtained in step S41. At this stage, different types of defects are displayed in layers and density statistics are performed, and then a wire interface defect distribution map reflecting the distribution law of various defects is constructed. In a specific implementation, different types of defects are classified by using a spatial clustering algorithm (such as K-means clustering or DBSCAN). According to the spatial position of each defect, the distribution density of different defects at the metal wire interface and in its depth direction is statistically calculated. By constructing a defect distribution map, the concentration and distribution trend of different types of defects in space can be intuitively seen. This process usually requires analyzing the category, location and defect density of the defect in the surrounding area to generate a map reflecting the defect distribution, such as a color-coded heat map, to identify defect-dense areas and defect-sparse areas. According to the spatial distribution law of different defects, engineers can evaluate which areas need special attention, which defect types are more common, and their potential impact on the overall wire quality.

[0131] Step S43: performing weighted calculation of the impact of various defects on wire performance based on the wire interface defect distribution map, evaluating the quality level of the wire as a whole and in local areas, and forming the wire interface defect severity;

[0132] According to the wire interface defect distribution map constructed in step S42, the embodiment of the present invention performs a weighted calculation of the impact of various defects on wire performance to evaluate the quality level of the wire as a whole and in local areas, and to form the severity of wire interface defects. In this process, it is first necessary to define the weight of the impact of each type of defect on the performance of the metal wire. For example, certain defects (such as through cracks and local peeling) may have a greater impact on the mechanical properties of the wire (such as tensile strength and impact resistance), while other types of defects (such as tiny pores) may have a smaller impact on the performance. By combining material science knowledge, a corresponding weight coefficient is assigned to each defect, and then a weighted calculation is performed based on the distribution of defects on the surface and inside of the metal wire to obtain the severity of the defect in each area. In this way, the quality level of the metal wire as a whole and in local areas can be evaluated, providing a quantitative basis for subsequent quality control and optimization.

[0133] Step S44: determining the characteristic pattern and severity of the current wire material defect according to the wire material interface defect severity, and generating a wire material defect rating result;

[0134] The embodiment of the present invention determines the characteristic pattern and severity of the current wire defect based on the severity of the wire interface defect, and generates a wire defect rating result. Specifically, based on the defect severity data obtained by the above calculation, a classification algorithm (such as a decision tree, support vector machine, etc.) is used to rate the defects of the wire. For example, if the defect severity of a certain section of wire exceeds a preset threshold, the section of wire may be marked as a "serious defect", while another section with fewer defects may be rated as a "minor defect". Through this rating method, the quality status of the wire in different production batches or under different process conditions can be fully understood, helping quality control personnel in the production process to discover and solve problems in a timely manner. The generated defect rating results will serve as standardized evaluation data for wire quality, providing support for subsequent product inspection, customer acceptance and quality traceability.

[0135] Step S45: Correlation analysis is performed on the wire defect rating results and the process parameter time series data recorded during the wire production process to identify the corresponding relationship between defect formation and changes in specific process parameters, thereby forming wire production process correlation data.

[0136] The embodiment of the present invention matches the process parameters (such as temperature, stretching rate, cooling rate, etc.) recorded in the production process with the defect data in a time series, and uses statistical analysis methods (such as correlation analysis or regression analysis) to explore the potential relationship between process parameters and defect type and defect severity. For example, when a certain production batch of wire is processed at a higher temperature, more pore defects may be generated, while a cooling rate that is too fast may lead to an increase in peeling defects. Through this correlation analysis, the key control parameters in the process can be identified, providing data support for the optimization of the production process. Ultimately, these process-related data will help the production team adjust the process flow, reduce the incidence of defects, and thus improve the overall quality and production efficiency of the wire.

[0137] Preferably, step S5 includes the following steps:

[0138] Step S51: constructing a process parameter optimization model based on the wire rod production process correlation data, identifying key process parameters that affect interface quality, and generating a process parameter priority list;

[0139] The embodiment of the present invention uses the wire production process associated data obtained in step S45 as input, and uses statistical analysis and machine learning algorithms (such as random forests, support vector machines, etc.) to identify key process parameters that affect interface quality. These process parameters may include temperature, stretching rate, cooling rate, pressure, etc. Through model training, it is possible to identify which process parameters have a significant impact on wire defect formation and interface quality, and sort the process parameters according to their degree of influence. By calculating the impact of each parameter on quality indicators such as defect type and defect severity, a process parameter priority list can be generated to help engineers focus on optimizing those parameters that have the greatest impact on quality, thereby improving the consistency and performance of wire products.

[0140] Step S52: designing a parameter adjustment plan according to the process parameter priority list, formulating a process parameter optimization strategy based on different types of interface defects, and generating process parameter adjustment data;

[0141] The embodiment of the present invention designs a targeted process parameter optimization strategy based on different types of interface defects (such as peeling, pores, inclusions, etc.) and their association with process parameters. For each defect type, the process parameter that has the greatest impact on it is selected and adjusted. For example, if it is found that the cooling rate has an important influence on the formation of pore defects, the cooling rate is adjusted. If the temperature has a greater impact on the peeling defect, the peeling defect is reduced by optimizing the temperature curve. The design of the optimization strategy must take into account the interaction of different process parameters in the production process. Therefore, the synergistic effect between the various parameters must be comprehensively considered to ensure that the adjustment plan can effectively control multiple defect types. Ultimately, the generated process parameter adjustment data contains the specific adjustment value and adjustment strategy for each parameter, providing a basis for subsequent process verification and implementation.

[0142] Step S53: using the process parameter adjustment data to perform production process simulation verification, predicting the effect of the parameter adjustment on the interface quality, and generating process optimization prediction data;

[0143] The embodiment of the present invention uses the process parameter adjustment data generated in step S52 to perform production process simulation verification. First, the effect of the adjusted process parameters on the interface quality is simulated through a computer simulation platform or a model based on historical production data. During the simulation process, based on different process parameter adjustment schemes, it is predicted whether the adjusted production process can effectively reduce specific types of interface defects, especially those defects that have a greater impact on wire performance. The effectiveness of the parameter adjustment is verified by comparing the defect distribution and defect severity before and after the simulation. The generated process optimization prediction data includes prediction results for the number, severity, distribution, etc. of defects. These data can provide a scientific basis for process adjustments in the actual production process, and help the process optimization team better understand the potential impact of parameter adjustments on wire quality.

[0144] Step S54: converting the process optimization prediction data into specific production equipment control parameters to generate wire manufacturing process optimization data;

[0145] The embodiment of the present invention converts the process optimization prediction data in step S53 into specific production equipment control parameters. Based on the process optimization prediction data, the optimal process parameters are selected and converted into specific production equipment control instructions. For example, the optimized parameters such as temperature, stretching rate, cooling rate, etc. are input into the control system of the production equipment. This process includes quantification and standardization of parameters and conversion into specific control values that can be executed by the production line. These optimization data will be directly applied to the automatic control system of the production equipment to ensure that the production process meets the optimal process requirements, so as to reduce the occurrence of defects and ensure the quality of the wire. The generated wire manufacturing process optimization data includes the control parameters of each production equipment and their variation range, which are used to achieve precise control of the wire quality during the production process.

[0146] Step S55: adjusting the wire production process parameters in real time based on the wire manufacturing process optimization data, collecting the process parameters during the production process, and generating process adjustment execution data;

[0147] The embodiment of the present invention is based on the wire manufacturing process optimization data generated in step S54, and at this stage, the wire production process parameters are adjusted in real time. By real-time monitoring of various process parameters on the production line and making real-time adjustments based on the optimization data, such as adjusting the parameters of the heating furnace temperature, stretching machine speed, cooling spray system, etc. During the adjustment process, the production system will automatically correct the parameters based on the process data collected in real time to ensure accurate control of the parameters in the production process. At the same time, various process parameters such as temperature, pressure, speed, tension, etc. need to be collected in real time during the production process. These data will be used for subsequent analysis to determine whether the adjusted process achieves the expected results. The generated process adjustment execution data includes the process parameter values and their execution status at each adjustment moment, providing basic data support for further analysis and optimization.

[0148] Step S56: performing interface quality inspection according to the process adjustment execution data to obtain wire interface performance evaluation data.

[0149] The embodiment of the present invention detects the interface quality of the wire based on the process adjustment execution data generated in step S55. Advanced non-destructive testing technologies (such as X-ray imaging, ultrasonic testing, laser scanning, etc.) are used to detect the interface quality of the wire adjusted during the production process to evaluate the type, quantity, size and distribution of its surface defects and internal defects. By comparing the changes in the type and number of defects, the improvement effect of the process parameter adjustment on the interface quality can be verified. The generated wire interface performance evaluation data includes the distribution and severity of various defects and the mechanical properties of the wire (such as tensile strength, flexural strength, etc.). These data will provide a quantitative evaluation of the process optimization effect and provide feedback for subsequent quality improvements to further adjust the production process.

[0150] The present invention further provides a metal wire data analysis system for executing the metal wire data analysis method described above, the metal wire data analysis system comprising:

[0151] The data acquisition and modeling module is used to obtain the wire interface data of the composite metal; collect the wire interface feature matrix based on the wire interface data; and construct the wire interface feature model based on the wire interface feature matrix;

[0152] The differential signal processing module is used to perform differential signal conversion processing on the wire interface feature model, extract the wire interface change characteristics based on the differential signal conversion results, and construct a preliminary indication map of wire interface defects based on the wire interface change characteristics;

[0153] The multi-scale decomposition module is used to perform multi-scale decomposition processing based on the preliminary indication map of the wire interface defect to obtain interface defect characteristic data; identify the type characteristics of the wire interface defect based on the interface defect characteristic data; match and correct the wire interface defect type characteristics with the preset physical constraint model to generate accurate defect characterization data;

[0154] The defect distribution module is used to establish a wire interface defect distribution map based on the precise defect characterization data; calculate the wire interface defect severity based on the wire interface defect distribution map; and analyze the wire production process related data based on the wire interface defect severity;

[0155] The quality verification module is used to generate wire manufacturing process optimization data based on wire production process related data; apply the wire manufacturing process optimization data to the wire production process parameter adjustment, and verify the wire interface quality to obtain wire interface performance evaluation data.

[0156] The present invention is therefore intended to be illustrative and non-restrictive in all respects, with the scope of the invention being defined by the appended claims rather than the foregoing description, and all changes that come within the meaning and range of equivalents of the application documents are intended to be embraced therein.

[0157] The foregoing description is intended only to provide specific embodiments of the present invention, which will enable those skilled in the art to understand and implement the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not intended to be limited to the embodiments shown herein, but is to be construed in the widest possible manner consistent with the principles and novel features disclosed herein.

Claims

1. A metal wire data analysis method, characterized in that: The following steps are involved: Step S1: obtaining wire interface data of the composite metal; Collecting a wire interface feature matrix based on wire interface data; and constructing a wire interface feature model based on the wire interface feature matrix; Step S2: performing differential signal conversion processing on the wire interface feature model, extracting wire interface change characteristics based on the differential signal conversion results; and constructing a preliminary wire interface defect indication map based on the wire interface change characteristics; Step S3: performing multi-scale decomposition processing based on the preliminary indication map of the wire interface defect to obtain interface defect characteristic data; Identify wire interface defect type characteristics based on interface defect characteristic data; Match and calibrate the wire interface defect type characteristics with the preset physical constraint model to generate accurate defect characterization data; Step S4: establishing a wire interface defect distribution map based on the defect precise characterization data; and calculating the wire interface defect severity based on the wire interface defect distribution map; Analyze wire production process-related data based on wire interface defect severity; Step S5: generating wire rod manufacturing process optimization data based on the wire rod production process associated data; The wire manufacturing process optimization data is applied to the adjustment of wire production process parameters, and the wire interface quality is verified to obtain the wire interface performance evaluation data.

2. The metal wire material data analysis method according to claim 1, characterized in that: Step S1 includes the following steps: Step S11: Setting up a wire detection device, fixing the composite metal wire to be tested within the detection area of the sensor array, thereby obtaining wire cross-sectional image data and surface feature data; Step S12: performing morphological feature reconstruction processing on the wire cross-section image data and the surface feature data, thereby obtaining wire interface data of the composite metal; Step S13: determining the ultrasonic emission position and parameters based on the wire interface data, transmitting a multi-band ultrasonic signal to the fixed composite metal wire, receiving the wire interface echo signal, and calculating the ultrasonic response parameters based on the echo signal intensity, phase, and delay time; Step S14: determining an eddy current detection path based on the wire interface data, scanning the wire using a multi-frequency eddy current sensor, measuring changes in the electromagnetic field response at different positions of the wire, and calculating electromagnetic permeability distribution data based on the changes in the electromagnetic field response; Step S15: setting X-ray scanning parameters based on the wire interface data, performing a tomographic scan of the wire using low-energy X-rays, and generating an interface density gradient map; Step S16: combining the ultrasonic response parameters, the electromagnetic permeability distribution data, and the interface density gradient map to obtain a wire interface characteristic matrix; Step S17: constructing a wire interface feature model according to the wire interface feature matrix.

3. The metal wire material data analysis method according to claim 2, characterized in that: Step S17 includes the following steps: Step S171: performing data standardization processing on the wire interface characteristic matrix, incorporating different physical quantities into a unified measurement system, and generating standardized interface characteristic data; Step S172: establishing a three-dimensional coordinate system based on the standardized interface feature data, mapping the wire geometry information into the spatial coordinate system, and forming wire interface spatial mapping data; Step S173: constructing a basic topological structure of the interface using the wire interface space mapping data, determining the continuity and morphological characteristics of the interface, and generating interface topological structure data; Step S174: analyzing the spatial distribution pattern of interface features based on the interface topology data, identifying the feature change pattern of the interface region, and forming interface region feature clustering data; Step S175: Screening key characteristic parameters sensitive to interface quality based on the interface region feature clustering data to obtain a set of interface key characteristic parameters; Step S176: constructing a feature vector space using the interface key feature parameter set, and performing dimensionality reduction processing to generate interface feature dimensionality reduction data; Step S177: establishing a wire interface mathematical model based on the interface feature dimensionality reduction data, and performing physical property constraints on the wire interface mathematical model to obtain a wire interface feature model.

4. The metal wire material data analysis method according to claim 3, characterized in that: Step S2 includes the following steps: Step S21: establishing a reference area of the wire interface feature model, selecting an interface section without obvious defects as a reference standard, and generating wire interface reference feature data; Step S22: performing point-to-point comparison between each detection area in the wire interface feature model and the wire interface reference feature data, calculating feature difference values, and forming original wire interface differential data; Step S23: performing statistical normalization processing on the original differential data of the wire interface and performing neighborhood enhancement processing to obtain enhanced differential data of the wire interface; Step S24: setting multi-level threshold conditions based on the wire interface enhanced differential data, and performing hierarchical screening on the differential signals to form wire interface differential signal conversion data; Step S25: Identifying region groups with similar change patterns based on the wire interface differential signal conversion data, thereby extracting wire interface change characteristics; Step S26: Perform spatial mapping based on the wire interface variation characteristics, determine the boundaries and ranges of high-confidence defect areas, and construct a preliminary wire interface defect indication map.

5. The metal wire material data analysis method according to claim 4, characterized in that: The multi-scale decomposition process based on the preliminary indication map of the wire interface defect in step S3 includes: Perform multi-resolution segmentation on the preliminary indication image of wire interface defects, divide the image into macro-scale, meso-scale and micro-scale detection areas, and generate multi-level defect decomposition data; Perform morphological expansion processing on the macro-scale detection area to enhance the continuity characteristics of the interface crack and generate macro-defect connected domain data; Adaptive threshold segmentation is performed on the mesoscopic detection area to separate the grain boundary slip band and the second phase precipitation characteristics, generating mesoscopic defect separation data; Perform Laplace sharpening filtering on the micro-scale inspection area to generate micro-defect enhancement data; Input multi-level defect decomposition data, macroscopic defect connected domain data, mesoscopic defect separation data, and microscopic defect enhancement data into a multi-channel data fusion network to generate cross-scale defect correlation data; Construct a three-dimensional defect space map based on cross-scale defect correlation data, mark the distribution density of defects in the X / Y / Z axes, and generate a defect space distribution vector; Perform principal component analysis on the defect spatial distribution vector to generate a defect core feature data set; The defect core feature dataset is matched with the preset defect type database by cosine similarity, and feature combinations with similarity > 85% are screened to generate interface defect feature data.

6. The metal wire material data analysis method according to claim 5, characterized in that: Identifying the wire interface defect type characteristics based on the interface defect characteristic data in step S3 includes: Extract the defect length, width, depth and area ratio characteristic values from the interface defect feature data to generate the defect basic geometric feature set; Compare the defect basic geometric feature set with the standard defect geometric template in the defect type database one by one to calculate the geometric similarity; Based on the geometric similarity, a candidate set of defect types with a matching degree of >75% is selected; Based on the defect type candidate set, the density distribution of defects in the material interface area, the spacing between adjacent defects, and the arrangement direction are calculated to generate the defect spatial distribution feature vector; Defect groups are divided based on density peak detection according to the defect spatial distribution feature vector to generate defect spatial clustering data; Extract defect edge morphological features based on defect spatial clustering data, quantify edge roughness, fractal dimension and curvature change rate, and generate defect edge feature parameter set; Perform secondary matching based on the defect edge feature parameter set and the defect type database to correct misjudgments in the preliminary defect classification data and generate optimized defect classification results; The optimized defect classification results are judged according to the preset defect type judgment rules to generate wire interface defect type features including defect type code, confidence level and position coordinates.

7. The metal wire material data analysis method according to claim 6, characterized in that: Matching and correcting the wire interface defect type characteristics with the preset physical constraint model in step S3 includes: Establish a physical constraint model library for wire interface defects, including the physical property parameters and constraints of delamination, pores and inclusions; Parametrically compare the wire interface defect type characteristics with the wire interface defect physical constraint model library to generate preliminary defect type matching data; Based on the preliminary matching data of the defect type, the physical rationality of the defect characteristics is evaluated, the constraint conditions are verified, and the physical rationality assessment of the defect is formed; Perform feature correction based on physical constraint rules on the physical rationality evaluation of defects to obtain defect feature correction data; Calculate the quantitative defect characterization parameters including defect size, depth, morphology and distribution characteristics based on the defect feature correction data to generate a set of defect quantitative characterization parameters; The defect quantitative characterization parameter set is matched with the defect evolution law in the wire interface defect physical constraint model library, and the defect formation mechanism and development trend are evaluated to generate accurate defect characterization data.

8. The metal wire material data analysis method according to claim 7, characterized in that: Step S4 includes the following steps: Step S41: spatially mapping the defect precise characterization data based on the axial and radial coordinates of the wire rod to generate wire rod defect spatial positioning data containing three-dimensional defect distribution information; Step S42: Based on the wire defect spatial positioning data, different types of defects are displayed in layers and density statistics are performed to construct a wire interface defect distribution map reflecting the distribution rules of various defects; Step S43: performing weighted calculation of the impact of various defects on wire performance based on the wire interface defect distribution map, evaluating the quality level of the wire as a whole and in local areas, and forming the wire interface defect severity; Step S44: determining the characteristic pattern and severity of the current wire material defect according to the wire material interface defect severity, and generating a wire material defect rating result; Step S45: Correlation analysis is performed on the wire defect rating results and the process parameter time series data recorded during the wire production process to identify the corresponding relationship between defect formation and changes in specific process parameters, thereby forming wire production process correlation data.

9. The metal wire material data analysis method according to claim 8, characterized in that: Step S5 includes the following steps: Step S51: constructing a process parameter optimization model based on the wire rod production process correlation data, identifying key process parameters that affect interface quality, and generating a process parameter priority list; Step S52: designing a parameter adjustment plan according to the process parameter priority list, formulating a process parameter optimization strategy based on different types of interface defects, and generating process parameter adjustment data; Step S53: using the process parameter adjustment data to perform production process simulation verification, predicting the effect of the parameter adjustment on the interface quality, and generating process optimization prediction data; Step S54: converting the process optimization prediction data into specific production equipment control parameters to generate wire manufacturing process optimization data; Step S55: adjusting the wire production process parameters in real time based on the wire manufacturing process optimization data, collecting the process parameters during the production process, and generating process adjustment execution data; Step S56: performing interface quality inspection according to the process adjustment execution data to obtain wire interface performance evaluation data.

10. A metal wire data analysis system, characterized in that: For executing the metal wire data analysis method according to claim 1, the metal wire data analysis system comprises: The data acquisition and modeling module is used to obtain the wire interface data of the composite metal; collect the wire interface feature matrix based on the wire interface data; and construct the wire interface feature model based on the wire interface feature matrix; The differential signal processing module is used to perform differential signal conversion processing on the wire interface feature model, extract the wire interface change characteristics based on the differential signal conversion results, and construct a preliminary indication map of wire interface defects based on the wire interface change characteristics; The multi-scale decomposition module is used to perform multi-scale decomposition processing based on the preliminary indication map of the wire interface defect to obtain interface defect characteristic data; identify the type characteristics of the wire interface defect based on the interface defect characteristic data; match and correct the wire interface defect type characteristics with the preset physical constraint model to generate accurate defect characterization data; The defect distribution module is used to establish a wire interface defect distribution map based on the precise defect characterization data; calculate the wire interface defect severity based on the wire interface defect distribution map; and analyze the wire production process related data based on the wire interface defect severity; The quality verification module is used to generate wire manufacturing process optimization data based on wire production process related data; apply the wire manufacturing process optimization data to the wire production process parameter adjustment, and verify the wire interface quality to obtain wire interface performance evaluation data.

Images