Method and system for detecting oxidation layer on surface of copper rod

Abstract

The invention relates to the technical field of copper rod oxide layer detection, in particular to a copper rod surface oxide layer detection method and system, and the method comprises the steps: building a calibration database, detecting a to-be-detected copper rod surface oxide layer, and collecting spectral image data and laser scattering data; based on the spectral image data, component parameters and thickness parameters are obtained, and based on the laser scattering data, scattered light intensity is analyzed to obtain scattering parameters; converting the component parameter, the thickness parameter and the scattering parameter into a component index, a thickness index and a compactness index by referring to the calibration database; a judgment rule comprising a plurality of judgment channels is constructed, the component index, the thickness index and the compactness index are judged through the judgment channels, and each judgment channel outputs an initial detection signal; and integrating all the initial detection signals to generate a detection result. According to the invention, the multi-dimensional characteristics of the copper rod oxide layer are effectively detected, the detection accuracy is improved, and a comprehensive and reliable detection result is obtained.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of copper rod oxidation layer detection, and particularly relates to a copper rod surface oxidation layer detection method and system. BACKGROUND

[0002] Copper rod is an important basic raw material in the power, electronics, communication and other industries, and its surface quality directly affects the success of subsequent wire drawing, tinning and other processing technologies and the performance of the final product. In the production and storage process, the surface of the copper rod will inevitably react with oxygen in the air to form an oxidation layer. The oxidation layer mainly contains cuprous oxide and copper oxide, significantly increasing the contact resistance in the subsequent processing process, causing abnormal wear of the wire drawing die and even wire breakage, and affecting the adhesion of the plated layer. Rapid and accurate detection and evaluation of the copper rod surface oxidation layer is a key link in quality control during production.

[0003] Currently, the detection of copper rod surface oxidation layer in industrial production usually adopts machine vision detection method or single spectrum analysis method. The machine vision method mainly responds to color changes, but color is easily disturbed by lighting conditions, camera performance and surface cleanliness, and cannot effectively distinguish between cuprous oxide and copper oxide with similar colors but different properties, nor can it obtain the thickness and compactness information of the oxidation layer. Although the single spectrum analysis method can provide certain composition and thickness information, it is not sensitive to the microstructure of the oxidation layer, and the microstructure is exactly the key factor affecting its conductivity and processing performance. Moreover, relying on a single data source for analysis is easily disturbed by environmental factors, resulting in misjudgment, and cannot provide a comprehensive detection report on composition, thickness and compactness.

[0004] Therefore, there is a need for a copper rod surface oxidation layer detection method and system that can overcome the above-mentioned shortcomings. SUMMARY

[0005] In view of at least one of the above technical problems, the present application provides a copper rod surface oxidation layer detection method and system, which adopts a multi-source data detection method combining spectral feature extraction and laser scattering analysis to realize integrated quantification and stable determination of the composition, thickness and compactness of the oxidation layer.

[0006] The present application provides a copper rod surface oxidation layer detection method, comprising the following steps:

[0007] S1: Establish a calibration database, detect the surface oxidation layer of the copper rod to be tested, and collect spectral image data and laser scattering data;

[0008] S2: extracting a first spectral feature located in a first preset wavelength interval and a second spectral feature located in a second preset wavelength interval based on the spectral image data, and obtaining a composition parameter and a thickness parameter according to a comparison relationship between the first spectral feature and the second spectral feature, analyzing a scattering light intensity based on the laser scattering data to obtain a scattering parameter;

[0009] S3: converting the composition parameter, the thickness parameter and the scattering parameter into a composition index, a thickness index and a compactness index by referring to the calibration database;

[0010] S4: constructing a judgment rule comprising a plurality of judgment channels, judging the composition index, the thickness index and the compactness index through the judgment channels respectively, and outputting an initial detection signal from each of the judgment channels;

[0011] S5: integrating all the initial detection signals to generate a detection result.

[0012] In some embodiments of the present application, the first spectral feature located in the first preset wavelength interval and the second spectral feature located in the second preset wavelength interval are extracted, specifically:

[0013] The first preset wavelength interval is used to extract the first spectral feature corresponding to the cuprous oxide characteristic absorption peak;

[0014] The second preset wavelength interval is used to extract the second spectral feature corresponding to the copper oxide characteristic absorption peak.

[0015] In some embodiments of the present application, the composition parameter and the thickness parameter are obtained according to the comparison relationship between the first spectral feature and the second spectral feature, including:

[0016] The first spectral feature and the second spectral feature are subjected to spectral intensity normalization processing respectively, the absorption peak difference and baseline ratio of the two are calculated, and the relative content of copper oxide and cuprous oxide is determined based on the absorption peak difference and baseline ratio to obtain the composition parameter;

[0017] A thickness absorption curve is constructed, and the thickness parameter is obtained according to the absorption peak difference.

[0018] In some embodiments of the present application, the judgment rule comprising a plurality of judgment channels includes:

[0019] The judgment rule includes a composition judgment channel, a thickness judgment channel and a compactness judgment channel;

[0020] The composition judgment channel is used to compare the composition index with a preset composition threshold value to determine the main composition category of the oxide layer;

[0021] The thickness determination channel is used for comparing the thickness index with a preset thickness threshold interval to determine the thickness of the oxide layer.

[0022] The compactness determination channel is used for comparing the compactness index with a preset compactness threshold to determine the uniformity and compactness of the oxide layer structure.

[0023] Each of the determination channels outputs the initial detection signal.

[0024] In some embodiments of the present application, the composition determination channel is used for comparing the composition index with a preset composition threshold to determine the main composition category of the oxide layer, including:

[0025] A copper oxide standard information set and a cuprous oxide standard information set are established.

[0026] The first spectral feature and the second spectral feature are subjected to difference operation on the wavelength axis to draw a spectral difference curve of the composition index.

[0027] The wavelength positions, amplitudes and adjacent intervals of a plurality of extreme points of the spectral difference curve are extracted to form a to-be-detected information set.

[0028] The to-be-detected information set is matched with the copper oxide standard information set and the cuprous oxide standard information set respectively, an optimal matching path is determined, and a total cost corresponding to the optimal matching path is calculated through the to-be-detected information set.

[0029] The main category of the oxide layer is determined according to the total cost.

[0030] The copper oxide standard information set and the cuprous oxide standard information set contain the wavelength positions, amplitudes and adjacent intervals matched with the to-be-detected information set point by point.

[0031] In some embodiments of the present application, the total cost corresponding to the optimal matching path is calculated through the to-be-detected information set, including:

[0032] The wavelength difference D λ (i) is calculated for all matching points based on the wavelength positions.

[0033] The wavelength difference .

[0034] The amplitude difference D A (i) is calculated for all matching points based on the amplitudes.

[0035] The amplitude difference .

[0036] The interval difference D Δ (i) is calculated for all matching points based on the adjacent intervals.

[0037] the interval difference ;

[0038] adding the wavelength difference, the amplitude difference and the interval difference of all matching points to obtain the total cost;

[0039] wherein, i is an extreme point position in the information group to be measured, λ i is the wavelength position of the extreme point position, A i is the amplitude of the extreme point position; j is a matching extreme point position in the copper oxide standard information group or the cuprous oxide standard information group, λ j is the wavelength position of the extreme point position, A j is the amplitude of the extreme point position; Δλ i is the adjacent interval of adjacent extreme point positions in the information group to be measured, Δλ j is the adjacent interval of the matching extreme point position in the standard information group; δ λ is a preset tolerance parameter of the wavelength position, and is a constant; δ A is a preset tolerance parameter of the amplitude, and is a constant; W λ (i) is a wavelength weight, W A (i) is an amplitude weight, W Δ (i) is an interval weight, and all are constants, and W λ (i) + W A (i) + W Δ (i) = 1.

[0040] In some embodiments of the present application, the total cost further comprises a global penalty:

[0041] when there is an extreme point position in the information group to be measured that fails to match in the standard information group, a penalty value of a missing point position is added;

[0042] when there is an extreme point position in the standard information group that fails to be matched by the information group to be measured, a penalty value of an inserted point position is added;

[0043] when matching across adjacent extreme point positions results in discontinuous matching, a penalty value of inconsistent trend is added;

[0044] the sum of the penalty values is added to the total cost as the global penalty.

[0045] In some embodiments of the present application, the thickness determination channel is used to compare the thickness index with a preset thickness threshold interval, comprising:

[0046] Comparing the thickness index with the boundary values of each preset thickness threshold interval in sequence, when the thickness index falls into a certain interval, output the thickness level signal corresponding to the interval;

[0047] When the thickness index is near the interval boundary, further compare the difference between the center values of adjacent intervals, and output the thickness level signal corresponding to the smaller difference.

[0048] In some embodiments of the application, further comprising step S6, auxiliary judgment is performed on the oxide layer:

[0049] A micro-area excitation signal is applied to the surface of the copper rod to be tested, and transient response data of the region is collected;

[0050] According to the transient response data, the physical properties of the oxide layer are analyzed to obtain the corresponding auxiliary judgment signal;

[0051] According to the auxiliary judgment signal and the detection result, a comprehensive detection result is obtained.

[0052] The application also provides a copper rod surface oxide layer detection system, comprising:

[0053] A collection and storage module is used to establish a calibration database, detect the surface oxide layer of the copper rod to be tested, and collect spectral image data and laser scattering data;

[0054] A parameter extraction module is used to extract a first spectral feature located in a first preset wavelength interval and a second spectral feature located in a second preset wavelength interval based on the spectral image data, and obtain a composition parameter and a thickness parameter according to the comparison relationship between the first spectral feature and the second spectral feature, and analyze the scattering light intensity based on the laser scattering data to obtain a scattering parameter;

[0055] An index conversion module is used to convert the composition parameter, the thickness parameter and the scattering parameter into a composition index, a thickness index and a compactness index by referring to the calibration database;

[0056] A judgment and detection module is used to construct a judgment rule containing multiple judgment channels, judge the composition index, the thickness index and the compactness index through the judgment channels, output an initial detection signal from each judgment channel, integrate all the initial detection signals, and generate a detection result.

[0057] The beneficial effects of the present application are: the present application can overcome the deficiency of the split information dimension of the traditional single detection method by synchronously acquiring the spectral image data and laser scattering data in the detection process, and establishing three types of judgment channels of composition, thickness and density, significantly improving the comprehensiveness and reliability of the detection results; the interference influence under the conditions of surface roughness, light disturbance, and copper oxide and cuprous oxide mixed phase is effectively weakened by using spectral differential feature matching, thickness absorption curve and scattering parameter joint judgment, so that the detection stability can still be maintained in the complex production line environment. BRIEF DESCRIPTION OF DRAWINGS

[0058] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or prior art description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments described in the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0059] Figure 1 The step schematic diagram of a copper rod surface oxide layer detection method in the embodiment of the present application;

[0060] Figure 2 The flowchart of obtaining the composition parameter and the thickness parameter according to the comparison relationship between the first spectral feature and the second spectral feature in the embodiment of the present application;

[0061] Figure 3 The step diagram of the composition judgment channel comparing the composition index with the preset composition threshold value to judge the main composition category of the oxide layer in the embodiment of the present application;

[0062] Figure 4 The step diagram of the thickness judgment channel comparing the thickness index with the preset thickness threshold value interval in the embodiment of the present application;

[0063] Figure 5 The step schematic diagram of the auxiliary judgment of the oxide layer in the embodiment of the present application. DETAILED DESCRIPTION

[0064] The technical solutions in the embodiments of the present application will be described clearly and completely in combination with the drawings in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments.

[0065] It should be understood that when an element as a layer, region or plate is referred to as being "on" another element, it can be directly on the other element or intervening elements can also be present. In addition, it should be understood that when an element is referred to as being "connected" to or "coupled" to another element, it can be directly connected or coupled to the other element or intervening elements can be present. As used herein, the term "and / or" includes any and all combinations of one or more of the associated listed items.

[0066] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used in the description of the application herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. As used herein, the term "and / or" includes any and all combinations of one or more of the associated listed items.

[0067] The application provides a copper rod surface oxide layer detection method as shown in the drawings, comprising the following steps: Figures 1 to 5

[0068] S1: Establishing a calibration database, detecting the surface oxide layer of the copper rod to be measured, collecting spectral image data and laser scattering data; using a hyperspectral camera as a collection tool, under the uniform illumination of a specific broadband light source, the hyperspectral camera continuously collects a hyperspectral image cube of the surface of the copper rod, which contains two-dimensional spatial information and one-dimensional spectral information as spectral image data; using a semiconductor laser as a light source, the light beam is collimated after expansion and is incident on the surface of the copper rod at a certain angle such as 45°, and a CCD detector is used to face the surface of the copper rod to be measured to receive the scattering spot image, the intensity distribution of the scattering light is directly related to the roughness and granularity of the surface microtopography; the establishment of the calibration database can be in a laboratory environment, a series of standard copper rod oxide samples with known composition, known thickness and known density determined by an ellipsometer are prepared, the spectral image data and laser scattering data of the standard samples are collected using the same equipment as above, and the feature parameters are extracted and associated with the known indicators to model, thereby establishing a calibration database for subsequent parameter conversion.

[0069] ​S2: Based on the spectral image data, a first spectral feature located in a first preset wavelength interval and a second spectral feature located in a second preset wavelength interval are extracted, and a composition parameter and a thickness parameter are obtained according to a comparison relationship between the first spectral feature and the second spectral feature; scattering parameters are obtained by analyzing the scattering light intensity based on the laser scattering data; after the collected laser scattering image is preprocessed, such as grayscale and denoising, the gray standard deviation and the average gray value of the entire light spot image are calculated, and the two characteristic values are input into a model trained based on a calibration database, so that the scattering parameters reflecting the compactness and uniformity of the oxidation layer are output; the stronger the scattering is and the more dispersed the light spot is, the higher the scattering parameter value is, indicating that the surface is rougher and the oxidation layer structure is more loose.

[0070] S3: The composition parameter, the thickness parameter and the scattering parameter are converted into a composition index, a thickness index and a compactness index by referring to the calibration database; after the key features are extracted from the spectral data, the key features are input into the corresponding mapping relationship in the database to obtain a standardized value, which reflects the ratio of cupric oxide to cuprous oxide in the oxidation layer; when the value is close to a certain standard sample, it indicates that the composition of the copper rod to be measured is similar to that of the standard sample; according to the overall absorption characteristics of the spectrum, combined with the thickness information of the existing samples in the database, a thickness index is obtained by interpolation or fitting method, which can reflect whether the oxidation layer to be measured is in the thin layer, the middle layer or the thick layer; the statistical characteristics of the scattering image are input into the database to obtain a compactness index, which can distinguish between the dense layer and the porous layer, so as to reflect the tightness of the oxidation layer structure.

[0071] S4: A judgment rule including multiple judgment channels is constructed, and the composition index, the thickness index and the compactness index are judged through the judgment channels respectively, and each judgment channel outputs an initial detection signal; the main composition category, the thickness grade and the compactness level of the oxidation layer are judged, so that the results in different dimensions do not interfere with each other, and the stability can be maintained in complex situations.

[0072] S5: All the initial detection signals are integrated to generate a detection result; the integration method can be simple weighting or fusion based on logic rules. The final result is output in the form of a comprehensive conclusion, for example, the oxidation layer is mainly cupric oxide, the thickness is moderate, and the compactness is poor; the overall state of the oxidation layer can be directly reflected, which is convenient for the operator to quickly judge whether the production process is reasonable or whether the annealing, cooling and other links need to be adjusted.

[0073] In the embodiment, the multi-source data fusion of the detected spectral features and laser scattering features is implemented, and the calibration database and the multi-channel judgment rule are supplemented, so that the three key indexes of the composition, the thickness and the compactness of the oxidation layer are quantified and comprehensively judged.

[0074] In some embodiments of the present application, as Figure 2As shown, the first spectral feature located in the first preset wavelength interval and the second spectral feature located in the second preset wavelength interval are extracted, specifically:

[0075] The first preset wavelength interval is used to extract the first spectral feature corresponding to the cuprous oxide characteristic absorption peak;

[0076] The second preset wavelength interval is used to extract the second spectral feature corresponding to the cupric oxide characteristic absorption peak.

[0077] The first preset wavelength interval is selected to cover the position of the cuprous oxide characteristic absorption peak, and the first spectral feature is extracted from the interval; the second preset wavelength interval covers the position of the cupric oxide characteristic absorption peak, and the second spectral feature is extracted from the interval; by directly capturing the differential signals exhibited by different oxides in spectral response, the problem of inaccurate determination caused by spectral data redundancy or interference components can be effectively avoided; by comparing and calculating the two spectral features, the specificity and accuracy of component analysis can be improved, and reliable reference for derivation of thickness parameters can be further provided; by using the characteristic peaks corresponding to different intervals as the basis for component determination, the sensitivity and distinguishability of the detection result to the real composition of the oxide layer can be significantly enhanced, so that the final output of the component index and the thickness index is more in line with the actual situation, and the accuracy and stability of the overall detection method are improved.

[0078] On the basis of the above embodiments, as shown in Figure 2 The component parameter and the thickness parameter are obtained according to the comparison relationship between the first spectral feature and the second spectral feature, including:

[0079] The first spectral feature and the second spectral feature are respectively subjected to spectral intensity normalization processing, the absorption peak difference and the baseline ratio of the two are calculated, and the relative content of cupric oxide and cuprous oxide is determined based on the absorption peak difference and the baseline ratio, so as to obtain the component parameter;

[0080] The thickness absorption curve is constructed, and the thickness parameter is obtained according to the absorption peak difference.

[0081] The first spectral feature and the second spectral feature are normalized respectively, so that the spectral intensity eliminates the influence brought by light source intensity, surface reflection condition or detection environment change, the absorption difference value of two characteristic peaks and the baseline ratio corresponding to the absorption difference value are calculated on the normalized spectral curve, so that a stable characteristic quantity related to the composition of the surface oxide layer of the copper rod to be measured is obtained, through comprehensive analysis of the absorption peak difference and the baseline ratio, the relative content relationship of cupric oxide and cuprous oxide can be effectively deduced, and then the composition parameter is obtained, and then the thickness absorption curve is constructed according to the corresponding relationship between the spectral absorption intensity and the thickness, and the thickness parameter of the oxide layer is determined by combining the position and amplitude of the absorption peak difference. Both the accuracy of the composition analysis in the complex detection environment and the deviation that may be caused by the single peak value method can be avoided, so that the finally obtained composition parameter and thickness parameter are more representative and stable, thereby providing reliable input data for subsequent index conversion and judgment.

[0082] The thickness absorption curve can establish a relationship model between the absorption intensity and the material thickness based on the Lambert-Beer law, calculate or simulate the absorption change of the spectrum under different thickness conditions, obtain the trend curve of the spectral characteristic peak difference with the thickness change, or prepare a series of copper rod oxide layer standard samples with different thicknesses, accurately measure the actual thickness by using a microscope section observation or a film thickness meter, and then directly draw the corresponding curve of the absorption peak difference and the thickness in combination with the spectral test results, so that the thickness absorption curve is highly consistent with the actual material state.

[0083] In some embodiments of the present application, a judgment rule comprising a plurality of judgment channels is constructed, including:

[0084] The judgment rule comprises a composition judgment channel, a thickness judgment channel and a compactness judgment channel;

[0085] The composition judgment channel is used for comparing the composition index with a preset composition threshold value to judge the main composition category of the oxide layer;

[0086] The thickness judgment channel is used for comparing the thickness index with a preset thickness threshold value interval to judge the thickness of the oxide layer;

[0087] The compactness judgment channel is used for comparing the compactness index with a preset compactness threshold value to judge the uniformity and compactness of the structure of the oxide layer;

[0088] Each judgment channel outputs an initial detection signal.

[0089] The composition index of the copper rod to be tested is input into a composition judgment channel, which compares the index with a preset composition threshold value, thereby judging the main composition category of the oxide layer, for example, whether the oxide layer is dominated by copper oxide or cuprous oxide; subsequently, the thickness index is input into a thickness judgment channel, which outputs a level signal of the oxide layer thickness being in a thin layer, a medium layer or a thick layer by comparison with a preset thickness threshold interval; and the compactness index is also input into a compactness judgment channel, which judges whether the structure of the oxide layer is uniform and compact by comparison with a compactness threshold value, and outputs corresponding level information.

[0090] The composition threshold value can be obtained by performing spectral testing on a series of standard samples with known composition proportions, establishing a corresponding relationship between the true analysis results of copper oxide and cuprous oxide contents and spectral characteristic indexes, and setting a critical value as the composition threshold value when the index value is in the transition zone between copper oxide and cuprous oxide, thereby distinguishing between the cases dominated by cuprous oxide and copper oxide; the thickness threshold interval can be obtained by preparing oxide layer samples with different thicknesses and performing spectral testing, corresponding the spectral absorption differences to the actual thicknesses, drawing a thickness absorption curve, and then dividing the thickness range into several intervals, for example, a thin layer, a medium layer and a thick layer, according to the actual needs of production and quality control, and taking the corresponding points on the curve as the interval boundaries, which reflect the physical correspondence between the spectral characteristics and the thickness, and meet the needs of rapid judgment in engineering detection; the compactness threshold value can usually be obtained through comparative experiments, that is, selecting control samples with dense and loose structures, collecting their scattering data and extracting the corresponding compactness indexes, and then determining the range of the compactness indexes and taking the boundary value between the two types of samples as the compactness threshold value; the threshold value can also be further corrected by statistical methods according to the actual detection needs, so as to ensure stability and consistency in batch detection.

[0091] Each judgment channel operates independently and outputs an initial detection signal, which can be a level, a category or a numerical interval, all of which provide support for subsequent comprehensive judgment. Through this multi-channel parallel judgment mode, the influence of a single index on the overall detection result can be avoided, and the final detection conclusion is more comprehensive, objective and stable.

[0092] On the basis of the above embodiment, as shown in Figure 3 The composition judgment channel is used for comparing the composition index with a preset composition threshold value and judging the main composition category of the oxide layer, including:

[0093] A copper oxide standard information group and a cuprous oxide standard information group are set up;

[0094] The first spectral feature and the second spectral feature are subjected to difference operation on the wavelength axis, and a spectral difference curve of the composition index is drawn;

[0095] extracting the wavelength position, amplitude and adjacent interval of the plurality of extreme points of the spectral difference curve to form a to-be-detected information group;

[0096] respectively matching the to-be-detected information group with the copper oxide standard information group and the cuprous oxide standard information group point by point, determining an optimal matching path, and calculating a total cost of the corresponding optimal matching path through the to-be-detected information group;

[0097] judging the main category of the oxide layer according to the total cost;

[0098] The copper oxide standard information group and the cuprous oxide standard information group contain the wavelength position, amplitude and adjacent interval matched with the to-be-detected information group point by point.

[0099] The copper oxide standard information group and the cuprous oxide standard information group are respectively established, and both of the information groups contain the extreme point position, amplitude size and distance between adjacent extreme points of a typical sample extracted in the spectral test process. In actual detection, the first spectral feature and the second spectral feature of the to-be-detected copper rod are subjected to difference operation on the wavelength axis to obtain a spectral difference curve of the composition index, and the wavelength position, amplitude value and adjacent interval of the plurality of extreme points on the curve are extracted to form a to-be-detected information group. The multi-dimensional feature information can be captured on a difference curve, instead of relying on a single peak value, thereby effectively enhancing the robustness of the judgment.

[0100] The to-be-detected information group is respectively matched with the copper oxide standard information group and the cuprous oxide standard information group point by point, and the differences in wavelength position, amplitude size and adjacent interval are considered in the matching process to find the optimal matching path between the standard information groups. After obtaining the optimal matching path, the corresponding total cost is further calculated based on the to-be-detected information group. The total cost is a quantitative index obtained by comprehensively calculating the differences of each matching point. The smaller the value is, the higher the similarity between the to-be-detected copper rod and a certain standard information group is. By comparing the total costs of the to-be-detected information group with the copper oxide standard information group and the cuprous oxide standard information group respectively, the main composition category of the oxide layer can be determined. For example, when the total cost of the copper oxide standard information group is significantly smaller than that of the cuprous oxide standard information group, it can be determined that the oxide layer is mainly copper oxide. Conversely, it is determined that it is mainly cuprous oxide.

[0101] On the basis of the above embodiment, the total cost of the corresponding optimal matching path is calculated through the to-be-detected information group, including:

[0102] calculating the wavelength difference D of all matching points based on the wavelength position λ (i),

[0103] wavelength difference ;

[0104] The peak position of the spectral curve is directly related to the chemical bond energy of the material. The peak positions of cuprous oxide and cupric oxide often have slight shifts. By calculating the wavelength difference D λ (i), the difference in chemical composition between the test information group and the standard information group can be directly reflected. The wavelength difference D λ (i) is calculated in the form of absolute value normalization, which ensures that the difference is always positive regardless of the direction of the shift, and physically corresponds to the degree of shift of the energy level structure. When the positions of the two extreme points on the wavelength axis are closer, the wavelength difference is smaller; otherwise, the difference is larger, and the wavelength weight W λ (i) is adjusted, which can adjust the contribution of the wavelength position to the total cost according to the importance in actual detection.

[0105] The amplitude difference D A (i) is calculated for all matching points based on the amplitude.

[0106] The amplitude difference D ;

[0107] The spectral amplitude is related to the absorption intensity of the material, reflecting the difference in concentration and thickness of the oxide layer, but the amplitude data is easily affected by noise such as light source intensity and detector sensitivity. The amplitude difference D A (i) highlights the sensitivity of small differences through the square term, so that slight composition changes can be captured; and the exponential decay factor suppresses the influence of large differences, avoiding the dominance of incidental noise such as detector saturation points in the determination; this combination can both amplify small differences and filter large abnormalities, equivalent to the combination of signal enhancement and noise suppression.

[0108] The distance difference D Δ (i) is calculated for all matching points based on the adjacent distance.

[0109] The distance difference D ;

[0110] The distance between adjacent extreme points reflects the overall structural characteristics of the spectrum, which is affected by the material lattice and electronic energy level structure. The relative distance is often more stable than a single peak and is not easily disturbed by systematic errors. The distance difference D Δ (i) is in the form of a relative ratio, because the spectral curve may shift as a whole, but the relative distance between adjacent peaks is stable. The ratio form can capture the inherent energy level structure characteristics of the material and has stronger discrimination.

[0111] The total cost is obtained by adding the wavelength difference, amplitude difference, and distance difference of all matching points.

[0112] where i is the extreme point in the test information group, λ i is the wavelength position of the extreme point, A iis the amplitude of the extreme point; j is the matched extreme point in the copper oxide standard information set or cuprous oxide standard information set, λ j is the wavelength position of the extreme point, A j is the amplitude of the extreme point; Δλ i is the adjacent interval of adjacent extreme points in the to-be-measured information set, Δλ j is the adjacent interval of the matched extreme point in the standard information set; δ λ is a preset tolerance parameter of the wavelength position, and is a constant; δ A is a preset tolerance parameter of the amplitude, and is a constant; W λ (i) is the wavelength weight, W A (i) is the amplitude weight, W Δ (i) is the interval weight, and W λ (i) + W A (i) + W Δ (i) = 1, and the weights can be optimized according to detection requirements, so that the method is suitable for both classification judgment and thickness and compactness analysis.

[0113] In spectral analysis, the differences between the to-be-measured information set and the standard information set mainly lie in three aspects: the shift of the wavelength position, the change of the chemical composition or the energy level structure causes the change of the peak position; the change of the amplitude, the absorption intensity is affected by the thickness and the composition concentration of the oxide layer, and the amplitude difference can reflect the difference in the content of the substance; the difference of the adjacent interval, the energy level interval of different materials is different, which changes the interval between peaks; therefore, in the calculation, the difference value is used to measure the deviation between two spectra, which is the most direct and most consistent with the physical actual comparison method, which can clearly quantify the deviation degree between the to-be-measured extreme point and the standard extreme point, and avoid ambiguous judgment.

[0114] The tolerance parameters δ λ and δ A can normalize the small deviation within a reasonable range, avoid misjudgment caused by amplification, reduce the influence of errors such as equipment resolution and environmental fluctuations during measurement, and enable the difference in different dimensions to be compared in the same scale; the tolerance parameters δ λ and δ A can be obtained in various ways, such as selecting a group of copper rod samples of known oxide layer types and thicknesses as calibration objects, obtaining the peak position and amplitude data of the samples through spectral measurement, analyzing the fluctuation ranges of these data under different experimental conditions (such as temperature, detection angle, and instrument resolution), and taking the statistical values of these fluctuation ranges as the tolerance parameters; or directly setting the tolerance parameters according to the instrument resolution, that is, δ λ is not less than the instrument resolution, and δ A is not less than the noise level of the system, so as to ensure that the judgment process remains stable within the normal measurement error range.

[0115] The total cost represents the comprehensive similarity between the to-be-tested information set and the standard information set, considers both local features (wavelength, amplitude) and overall features (interval), is sensitive to small differences and can avoid noise influence, provides a unified evaluation index, and the optimal matching path can be selected by comparing the total cost, and when the total cost is the smallest, it indicates that the similarity between the to-be-tested information set and a certain standard information set is the highest, and the corresponding standard information set is the determined main component category of the oxide layer.

[0116] On the basis of the above embodiment, the total cost further includes a global penalty:

[0117] When there is an extreme value point in the to-be-tested information set that fails to be matched in the standard information set, a penalty value of a missing point is added; if some extreme value points exist in the to-be-tested curve but do not exist in the standard curve, it indicates that the curve contains extra spectral features, and such a difference may mean that the material components are different.

[0118] When there is an extreme value point in the standard information set that fails to be matched in the to-be-tested information set, a penalty value of an inserted point is added; if the standard curve has an extreme value point of a characteristic peak, and the to-be-tested curve does not have an extreme value point of a corresponding peak, it indicates that the to-be-tested material lacks the key component of the standard substance, and the inserted point penalty ensures that such a missing feature is counted in the difference.

[0119] When matching across adjacent extreme value points leads to discontinuous matching, a penalty value of inconsistent trend is added; spectral extreme value points are usually arranged in order of energy level structure, and if matching appears to jump or be discontinuous, it indicates that there is a significant difference in the overall form between the to-be-tested curve and the standard curve, and the penalty of inconsistent trend is used to prevent the algorithm from pursuing local optimization and ignoring overall consistency.

[0120] The sum of the penalty values is added to the total cost as a global penalty, which ensures that the matching path can accurately reflect local differences and maintain global continuity and trend consistency, thereby improving the stability and reliability of the determination of the main component category.

[0121] In some embodiments of the present application, as shown in FIG. 1, Figure 4 the thickness determination channel is configured to compare the thickness index with preset thickness threshold intervals, and includes:

[0122] The thickness index is compared with the boundary values of each preset thickness threshold interval in sequence, and when the thickness index falls into a certain interval, a thickness level signal corresponding to the interval is outputted;

[0123] When the thickness index is near the boundary of an interval, the difference between the thickness index and the center value of the adjacent interval is further compared, and a thickness level signal corresponding to the smaller difference is outputted.

[0124] The thickness index is compared sequentially with the boundary values ​​of each preset thickness threshold interval. When the thickness index falls into a certain interval, the thickness level signal corresponding to that interval is output. Furthermore, when the thickness index is near the interval boundary, to avoid misjudgment and jump errors, this embodiment preferably introduces a secondary comparison mechanism. The difference between the thickness index and the center value of the adjacent interval is calculated separately, and the one with the smaller difference is selected as the final level determination basis, thereby outputting the corresponding thickness level signal. This can quickly give the thickness level through interval division, improving the efficiency of the determination. Introducing secondary comparison in the interval with ambiguous boundaries can effectively reduce misjudgment caused by critical point fluctuations, making the thickness determination smoother and more stable. The obtained thickness level signal is not only intuitive and easy to understand, but also provides a clear thickness grading reference for subsequent comprehensive detection results.

[0125] In some embodiments of the present invention, such as Figure 5 As shown, it also includes step S6, which assists in the determination of the oxide layer:

[0126] A micro-region excitation signal is applied to the surface of the copper rod under test, and transient response data of the region is collected.

[0127] Based on transient response data, the physical properties of the oxide layer are analyzed to obtain corresponding auxiliary judgment signals;

[0128] Based on the auxiliary judgment signal and the detection results, a comprehensive detection result is obtained.

[0129] The excitation signal can be transient thermal excitation generated by low-power laser pulses or electrochemical excitation generated by a small alternating voltage, and the intensity of the excitation signal is accurately controlled to ensure that the surface of the copper rod is not damaged; the transient response of the micro area under the action of the excitation signal is collected through a high-speed infrared thermal imager or a high-speed electrochemical impedance testing device; when the thermal excitation mode is used, the temperature decay curve during cooling is collected; when the electrochemical excitation mode is used, the corresponding impedance spectrum changes with time are collected; the collected transient response data is analyzed to extract parameters reflecting the characteristics of the oxide layer; in the case of thermal excitation, the thermal conductivity and thickness of the oxide layer are inferred through the slope and decay rate of the cooling curve; in the case of electrochemical excitation, the resistance and capacitance characteristics of the oxide layer are obtained through the equivalent circuit model fitting of the impedance spectrum, thereby reflecting the compactness and uniformity thereof; the corresponding auxiliary judgment signal is output, reflecting the thickness grade, compactness grade of the oxide layer, or the modification opinion of the main component classification of the oxide layer; the auxiliary judgment signal is compared with the detection result based on the spectrum and scattering, and when the two are consistent, the detection result is confirmed; when there is a difference between the two, the detection result is modified according to the auxiliary judgment signal, thereby obtaining the final comprehensive detection result; it can adapt to various interference factors in complex production environments, realize multi-angle verification of the thickness, compactness and composition of the oxide layer, avoid the limitations of a single method, and make the final judgment closer to the real state. Of course, the above embodiments only list the implementation modes of thermal excitation or electrochemical excitation, and the auxiliary excitation can also be extended to different probes and the like for implementation.

[0130] The application further provides a copper rod surface oxide layer detection system, comprising:

[0131] A collection and storage module is configured to establish a calibration database, detect the surface oxide layer of a copper rod to be tested, and collect spectral image data and laser scattering data.

[0132] A parameter extraction module is configured to extract a first spectral feature in a first preset wavelength interval and a second spectral feature in a second preset wavelength interval based on the spectral image data, and obtain a composition parameter and a thickness parameter according to a comparison relationship between the first spectral feature and the second spectral feature, and analyze the scattering light intensity based on the laser scattering data to obtain a scattering parameter.

[0133] An index conversion module is configured to convert the composition parameter, the thickness parameter and the scattering parameter into a composition index, a thickness index and a compactness index by referring to the calibration database.

[0134] A judgment and detection module is configured to construct a judgment rule comprising a plurality of judgment channels, judge the composition index, the thickness index and the compactness index through the judgment channels, output an initial detection signal from each judgment channel, integrate all the initial detection signals, and generate a detection result.

[0135] The above detection system in the application can effectively realize a copper rod surface oxide layer detection method, and the technical effects are as described in the above embodiment, which will not be repeated here.

[0136] Those skilled in the art should understand that the application is not limited to the above-mentioned embodiments, and the above-mentioned embodiments and descriptions in the specification are only to illustrate the principles of the application. Without departing from the spirit and scope of the application, various changes and improvements can be made to the application, and these changes and improvements all fall within the scope of the claimed application. The scope of protection of the application is defined by the appended claims and their equivalents.

Claims

1. A method for detecting the oxide layer on the surface of a copper rod, characterized in that, Includes the following steps: S1: Establish a calibration database, detect the oxide layer on the surface of the copper rod to be tested, and collect spectral image data and laser scattering data; S2: Based on the spectral image data, extract the first spectral feature located in the first preset wavelength range and the second spectral feature located in the second preset wavelength range, and obtain the composition parameter and thickness parameter according to the comparison relationship between the first spectral feature and the second spectral feature. Analyze the scattered light intensity based on the laser scattering data to obtain the scattering parameter. S3: Referring to the calibration database, convert the composition parameters, the thickness parameters, and the scattering parameters into composition indices, thickness indices, and density indices; S4: Construct a judgment rule containing multiple judgment channels, and use the judgment channels to judge the composition index, thickness index and density index respectively, and output an initial detection signal for each judgment channel; S5: Integrate all the initial detection signals to generate the detection results.

2. The method for detecting the oxide layer on the surface of a copper rod according to claim 1, characterized in that, Extracting the first spectral feature located in the first preset wavelength range and the second spectral feature located in the second preset wavelength range, specifically: The first preset wavelength range is used to extract the first spectral features corresponding to the characteristic absorption peak of cuprous oxide; The second preset wavelength range is used to extract the second spectral feature corresponding to the characteristic absorption peak of copper oxide.

3. The method for detecting the oxide layer on the surface of a copper rod according to claim 2, characterized in that, The composition parameters and thickness parameters are obtained based on the comparison between the first spectral features and the second spectral features, including: The first spectral feature and the second spectral feature are respectively normalized in terms of spectral intensity. The ratio of the difference in absorption peaks between the two to the baseline is calculated. Based on the ratio of the difference in absorption peaks to the baseline, the relative contents of copper oxide and cuprous oxide are determined to obtain the component parameters. Construct a thickness absorption curve and obtain the thickness parameter based on the difference in absorption peaks.

4. The method for detecting the oxide layer on the surface of a copper rod according to claim 1, characterized in that, Construct a judgment rule that includes multiple judgment channels, including: The judgment rules include a composition judgment channel, a thickness judgment channel, and a density judgment channel; The component determination channel is used to compare the component index with a preset component threshold to determine the main component category of the oxide layer; The thickness determination channel is used to compare the thickness index with a preset thickness threshold range to determine the oxide layer thickness. The density determination channel is used to compare the density index with a preset density threshold to determine the uniformity and density of the oxide layer structure. Each of the aforementioned determination channels outputs a copy of the initial detection signal.

5. The method for detecting the oxide layer on the surface of a copper rod according to claim 4, characterized in that, The component determination channel is used to compare the component index with a preset component threshold to determine the main component category of the oxide layer, including: Establish a copper oxide standard information group and a cuprous oxide standard information group; The difference between the first spectral feature and the second spectral feature is calculated on the wavelength axis to plot the spectral difference curve of the component index; The wavelength position, amplitude, and adjacent spacing of multiple extreme points of the spectral difference curve are extracted to form a group of information to be measured; The test information group is matched point by point with the copper oxide standard information group and the cuprous oxide standard information group to determine the optimal matching path, and the total cost of the corresponding optimal matching path is calculated through the test information group. The main category of the oxide layer is determined based on the total cost. The copper oxide standard information group and the cuprous oxide standard information group contain the wavelength position, amplitude and adjacent spacing that are matched point by point with the information group to be tested.

6. The method for detecting the oxide layer on the surface of a copper rod according to claim 5, characterized in that, The total cost corresponding to the optimal matching path is calculated using the group of information to be tested, including: Based on the wavelength position, the wavelength difference D is calculated for each matching point. λ (i), The wavelength difference ; Based on the amplitude, the amplitude difference D is calculated for each matching point. A (i), The amplitude difference ; Based on the adjacent spacing, the spacing difference D is calculated for each matching point. Δ (i), The distance difference ; The total cost is obtained by adding the wavelength difference, amplitude difference, and spacing difference of all matching points. Where i is the extreme point in the information group to be tested, and λ i For the wavelength position of this extreme point, A i The amplitude of the extreme point; j is the extreme point matched in the copper oxide standard information group or the cuprous oxide standard information group, λ j For the wavelength position of this extreme point, A j The amplitude at that extreme point; Δλ i Δλ is the adjacent distance between adjacent extreme points in the group of information to be measured. j The adjacent spacing of the extreme point positions matched in the standard information group; δ λ The tolerance parameter for the preset wavelength position is a constant; δ A The preset tolerance parameter for the amplitude is a constant; W λ (i) represents the wavelength weight, W A (i) represents the magnitude weight, W Δ (i) represents the spacing weights, all of which are constants, and W λ (i)+W A (i)+W Δ (i)=1.

7. The method for detecting the oxide layer on the surface of a copper rod according to claim 6, characterized in that, The total cost also includes a global penalty: For each extreme point in the information group to be tested that fails to match the standard information group, the penalty value for the missing point is increased by one. For each extreme point in the standard information group that fails to match the information group under test, the penalty value of the insertion point is increased by one. When a match crosses adjacent extreme points, causing the match to be discontinuous, add a penalty value for inconsistent trends. The sum of the penalty values ​​is included in the total cost as the global penalty.

8. The method for detecting the oxide layer on the surface of a copper rod according to claim 4, characterized in that, The thickness determination channel is used to compare the thickness index with a preset thickness threshold range, including: The thickness index is compared with the boundary values ​​of each preset thickness threshold range in turn. When the thickness index falls into a certain range, the thickness level signal corresponding to that range is output. When the thickness index is near the interval boundary, it is further compared with the difference of the center value of the adjacent interval, and the thickness level signal corresponding to the one with smaller difference is output.

9. The method for detecting the oxide layer on the surface of a copper rod according to claim 1, characterized in that, It also includes step S6, which assists in the determination of the oxide layer: A micro-region excitation signal is applied to the surface of the copper rod under test, and transient response data of the region is collected. Based on the transient response data, the physical properties of the oxide layer are analyzed to obtain the corresponding auxiliary judgment signal; Based on the auxiliary judgment signal and the detection result, a comprehensive detection result is obtained.

10. A copper rod surface oxide layer detection system, characterized in that, include: The data acquisition and storage module establishes a calibration database, detects the oxide layer on the surface of the copper rod under test, and acquires spectral image data and laser scattering data. The parameter extraction module extracts a first spectral feature located in a first preset wavelength range and a second spectral feature located in a second preset wavelength range based on the spectral image data. It then obtains composition parameters and thickness parameters based on the comparison relationship between the first spectral feature and the second spectral feature. Finally, it analyzes the scattered light intensity based on the laser scattering data to obtain scattering parameters. The index conversion module converts the composition parameters, thickness parameters, and scattering parameters into composition indices, thickness indices, and density indices by referring to the calibration database; The judgment and detection module constructs judgment rules containing multiple judgment channels. The judgment channels are used to judge the composition index, thickness index and density index respectively. Each judgment channel outputs an initial detection signal. All the initial detection signals are integrated to generate a detection result.

Images