Electrolytic copper foil warpage continuous test method and system

Abstract

This invention relates to the field of electrolytic copper foil testing technology, specifically disclosing a method and system for continuous testing of electrolytic copper foil warpage. The invention involves periodically cutting electrolytic copper foil samples; adjusting the testing environment and performing flattening, positioning, and benchmark calibration on the samples; performing three-dimensional panoramic scanning and multi-point scanning supplementation on the samples; reconstructing a three-dimensional surface model and comparing it against a benchmark to calculate multiple key warpage indicators; judging the pass / fail status and trend analysis of these key warpage indicators; generating a warpage test report; and implementing feedback control for continuous testing. This invention automatically generates three-dimensional surface diagrams and distribution curve visualization reports, intuitively presenting the defect evolution pattern; and dynamically adjusting electroplating parameters by triggering a process feedback control module, forming a test-analysis-control closed loop to suppress warpage degradation trends from the source, thereby improving product yield and production continuity.

Description

Technical Field

[0001] This invention belongs to the field of electrolytic copper foil testing technology, and particularly relates to a method and system for continuous testing of the warpage of electrolytic copper foil. Background Technology

[0002] Electrolytic copper foil testing is a systematic process of detecting and evaluating the performance and quality of copper foil materials produced by electrolysis using multiple dimensions, including physical, chemical, mechanical, and electrical methods. The purpose of the testing is to comprehensively understand key indicators such as thickness uniformity, surface flatness, tensile strength, elongation, roughness, resistivity, warpage, and heat resistance of the electrolytic copper foil, thereby ensuring its reliability and stability in downstream applications.

[0003] In the existing technology, the warpage test of electrolytic copper foil has the drawbacks of low level of intelligence and automation. Multiple processes require manual intervention, resulting in low testing efficiency and easy human error. In addition, the testing process is not continuous, making it difficult to detect the warpage trend in a timely manner. Summary of the Invention

[0004] The purpose of this invention is to provide a method and system for continuous testing of the warpage of electrolytic copper foil, aiming to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the embodiments of the present invention provide the following technical solutions:

[0006] A continuous testing method for the warpage of electrolytic copper foil, the method specifically includes the following steps:

[0007] Electrolytic copper foil samples are periodically cut, sample identification information is recorded, and the samples are transported to the warpage test channel.

[0008] Based on the sample identification information, the test environment is adjusted, and the electrolytic copper foil sample is flattened, positioned, and calibrated.

[0009] The electrolytic copper foil sample was subjected to three-dimensional panoramic scanning and multi-point scanning supplementation, and the data was spatiotemporally aligned and denoised to obtain sample warping data.

[0010] Based on the warping data of the sample, a three-dimensional surface model is reconstructed, and a benchmark comparison is performed to calculate multiple key warping indicators.

[0011] The system performs qualification assessment and trend analysis on multiple key warpage indicators, generates a warpage test report, and implements feedback control for continuous testing.

[0012] This invention also proposes a continuous testing system for the warpage of electrolytic copper foil, the system comprising:

[0013] The copper foil sample processing unit is used to periodically cut electrolytic copper foil samples, record sample identification information, and transport them to the warpage test channel.

[0014] The test environment adjustment unit is used to adjust the test environment according to the sample identification information, and to perform flat positioning and benchmark calibration on the electrolytic copper foil sample.

[0015] The three-dimensional scanning processing unit is used to perform three-dimensional panoramic scanning and multi-point scanning supplementation on the electrolytic copper foil sample, and to perform spatiotemporal alignment and noise reduction processing on the data to obtain sample warping data.

[0016] The model benchmark comparison unit is used to reconstruct a three-dimensional surface model based on the warping data of the sample, perform benchmark comparison, and calculate multiple key warping indicators.

[0017] The test report generation unit is used to perform qualification judgment and trend analysis on multiple key warpage indicators, generate a warpage test report, and perform feedback control for continuous testing.

[0018] Compared with the prior art, the beneficial effects of the present invention are:

[0019] 1. This invention achieves traceability and automated circulation of electrolytic copper foil samples by periodically cutting samples and associating them with production process parameters, significantly improving the representativeness and traceability efficiency of test samples; by adjusting the test environment and constructing an anti-interference benchmark model, it effectively overcomes optical measurement errors on metal surfaces, ensuring high precision and stability of benchmark calibration; by integrating three-dimensional panoramic scanning and dynamically supplemented scanning data, combined with a self-optimizing spatiotemporal alignment algorithm, it comprehensively captures the microscopic to macroscopic deformation characteristics of the sample surface, providing a high-fidelity data foundation without omissions for warpage analysis.

[0020] 2. This invention utilizes cross-scale surface reconstruction technology to achieve a smooth transition of macroscopic surfaces while preserving the sharpness of micro-wrinkles, thus constructing a traceable three-dimensional surface model. Through process decoupling analysis, it removes interfering factors such as electroplating current density and temperature, accurately extracting the intrinsic warping characteristics of the material. Through dynamic mesh comparison and multi-dimensional index calculation, it simultaneously obtains the maximum warping height, distribution curvature, and asymmetry index, comprehensively quantifying the warping state and locking the spatial distribution of defects.

[0021] 3. This invention uses a correlation process threshold library to judge the passability of warpage indicators in real time and predicts potential risks by combining historical data trends; it automatically generates three-dimensional surface diagrams and distribution curve visualization reports to intuitively present the evolution of defects; and it dynamically adjusts electroplating parameters by triggering the process feedback control module to form a test-analysis-control closed loop, thereby suppressing the warpage deterioration trend from the source and improving product yield and production continuity. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention.

[0023] Figure 1 A flowchart of the method provided by an embodiment of the present invention is shown.

[0024] Figure 2 A flowchart illustrating the process of cutting an electrolytic copper foil sample in the method provided in an embodiment of the present invention is shown.

[0025] Figure 3 A flowchart of the leveling positioning and reference calibration process in the method provided by an embodiment of the present invention is shown.

[0026] Figure 4 A flowchart illustrating the method for obtaining sample warpage data provided in an embodiment of the present invention is shown.

[0027] Figure 5 A flowchart illustrating the generation of a warpage test report in the method provided by an embodiment of the present invention is shown.

[0028] Figure 6 An application architecture diagram of the system provided in an embodiment of the present invention is shown.

[0029] Figure 7 A structural block diagram of the copper foil sample processing unit in the system provided in an embodiment of the present invention is shown.

[0030] Figure 8 A structural block diagram of the test environment conditioning unit in the system provided by an embodiment of the present invention is shown.

[0031] Figure 9 The diagram shows the structural block diagram of the test report generation unit in the system provided by an embodiment of the present invention. Detailed Implementation

[0032] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0033] Understandably, existing technologies for testing the warpage of electrolytic copper foil suffer from low levels of intelligence and automation. Multiple processes require manual intervention, resulting in low testing efficiency and a high risk of human error. Furthermore, the testing process is discontinuous, making it difficult to detect warpage trends in a timely manner.

[0034] To address the aforementioned issues, this invention involves periodically cutting electrolytic copper foil samples, recording their identification information, and transporting them to a warpage testing channel. Based on the sample identification information, the testing environment is adjusted, and the electrolytic copper foil samples are flattened, positioned, and calibrated. The electrolytic copper foil samples undergo three-dimensional panoramic scanning and multi-point scanning supplementation, with spatiotemporal alignment and noise reduction of the data to obtain sample warpage data. Based on the sample warpage data, a three-dimensional surface model is reconstructed, and benchmark comparisons are performed to calculate multiple key warpage indicators. These key warpage indicators are then assessed for compliance and trends, generating a warpage test report and implementing feedback control for continuous testing.

[0035] This invention achieves traceability and automated circulation of electrolytic copper foil samples by periodically cutting samples and associating them with production process parameters, significantly improving the representativeness and traceability efficiency of test samples. By adjusting the test environment and constructing an anti-interference benchmark model, it effectively overcomes optical measurement errors on metal surfaces, ensuring high accuracy and stability of benchmark calibration. By integrating three-dimensional panoramic scanning and dynamically supplemented scanning data, combined with a self-optimizing spatiotemporal alignment algorithm, it comprehensively captures the microscopic to macroscopic deformation characteristics of the sample surface, providing a high-fidelity data foundation without omissions for warpage analysis.

[0036] Figure 1 A flowchart of the method provided by an embodiment of the present invention is shown.

[0037] Specifically, the method for continuous testing of warpage of electrolytic copper foil includes the following steps:

[0038] Step S101: Periodically cut electrolytic copper foil samples, record sample identification information, and transport them to the warpage test channel.

[0039] In this embodiment of the invention, before the formal production of electrolytic copper foil, the copper foil sampling cycle and copper foil sampling parameters are determined. According to the copper foil sampling cycle and copper foil sampling parameters, electrolytic copper foil samples are periodically cut on the electrolytic copper foil production line. At the same time as the electrolytic copper foil samples are cut, a copper foil sample ID is generated, and the production batch, production time and process parameters corresponding to the electrolytic copper foil sample are obtained. By comprehensively organizing the copper foil sample ID, production batch, production time and process parameters, the sample identity information is recorded, and then the electrolytic copper foil sample is transported to the warpage test channel.

[0040] Specifically, Figure 2 A flowchart illustrating the process of cutting an electrolytic copper foil sample in the method provided in an embodiment of the present invention is shown.

[0041] In a preferred embodiment of the present invention, the periodic cutting of electrolytic copper foil samples, recording of sample identification information, and transport to the warpage test channel specifically includes the following steps:

[0042] Step S1011: Determine the copper foil sampling cycle and copper foil sampling parameters;

[0043] Step S1012: According to the copper foil sampling cycle and the copper foil sampling parameters, electrolytic copper foil samples are periodically cut on the electrolytic copper foil production line.

[0044] Step S1013: Generate copper foil sample ID;

[0045] Step S1014: Obtain the production batch, production time, and process parameters corresponding to the electrolytic copper foil sample;

[0046] Step S1015: Combine the copper foil sample ID, the production batch, the production time, and the process parameters to record the sample identification information;

[0047] Step S1016: The electrolytic copper foil sample is transported to the warpage test channel.

[0048] Furthermore, the electrolytic copper foil warpage continuity test method also includes the following steps:

[0049] Step S102: Adjust the test environment according to the sample identification information, and perform flat positioning and benchmark calibration on the electrolytic copper foil sample.

[0050] In this embodiment of the invention, test environment data is extracted from the process parameters of the sample identification information, and a reference plate is determined. According to the test environment data, the test environment temperature and humidity are adjusted. Then, on the reference plate, the electrolytic copper foil sample is flattened and positioned by electrostatic adsorption or vacuum pressing. The initial surface of the flattened and positioned electrolytic copper foil sample is then optically calibrated to construct a reference three-dimensional model, providing reference data for subsequent analysis.

[0051] Specifically, Figure 3 A flowchart of the leveling positioning and reference calibration process in the method provided by an embodiment of the present invention is shown.

[0052] In a preferred embodiment of the present invention, the step of adjusting the test environment according to the sample identification information and performing flattening, positioning, and benchmark calibration on the electrolytic copper foil sample specifically includes the following steps:

[0053] Step S1021: Extract test environment data from the process parameters of the sample identity information to determine the reference plate.

[0054] Step S1022: Adjust the test environment according to the test environment data;

[0055] Step S1023: The electrolytic copper foil sample is flattened and positioned on the reference plate.

[0056] Step S1024: Perform optical benchmark calibration on the flat and positioned electrolytic copper foil sample to construct a benchmark three-dimensional model.

[0057] In a preferred embodiment of the present invention, the optical reference calibration of the flat and positioned electrolytic copper foil sample and the construction of a reference three-dimensional model specifically include the following steps:

[0058] Step S10241: Deploy an adjustable spectral light source array on the flat and positioned surface of the electrolytic copper foil sample, and match the light intensity ratio of the blue light band and the infrared band according to the material reflection characteristics of the electrolytic copper foil sample. Simultaneously activate the ring polarization filter and diffuse blue light in the light source to suppress glare on the metal surface and generate an anti-interference composite lighting field.

[0059] Step S10242: In the anti-interference composite illumination field, a piezoelectric micro-displacement driven laser interferometer is used to generate a dense positioning grating on the edge of the reference plate, and the aerosol nozzle is controlled to spray transient fluorescent marker array in the non-detection area of ​​the sample to form a dual-source reference coordinate network.

[0060] Step S10243: Based on the spatiotemporal correlation characteristics of the dual-source reference coordinate network, drive the industrial camera to capture the surface deformation sequence, and start the line structured light scanner to acquire height topology data. Simultaneously record the spatial mapping relationship between the flat grating and the sample marker points to obtain the original morphology dataset.

[0061] Step S10244: Perform three-dimensional feature fusion on the original morphology dataset to generate a fused feature model with spatial constraints; wherein, performing three-dimensional feature fusion on the original morphology dataset includes: aligning the micro-deformation trajectories of the sequence images using an optical flow algorithm; fusing height topology data to construct an initial surface point cloud; and calibrating the overall offset of the sample using flat grating coordinates;

[0062] Step S10245: Perform physical property decomposition on the spatially constrained fusion feature model to obtain the artifact-free intrinsic morphology point cloud; wherein, the physical property decomposition on the spatially constrained fusion feature model includes: removing environmental temperature drift artifacts based on the material's thermal expansion coefficient; acquiring plate deformation monitoring data and eliminating reference surface errors through the plate deformation monitoring data; and preserving the inherent micromorphological features of the sample.

[0063] Step S10246: Layer the artifact-free intrinsic morphology point cloud to generate a baseline 3D model for error tracing; wherein, layering the artifact-free intrinsic morphology point cloud includes: embedding the ideal plane equation of the reference plate in the bottom layer; establishing the sample coordinate system transformation matrix in the middle layer; and integrating micro-morphology feature vectors in the surface layer.

[0064] Step S10247: Associate the baseline 3D model for error tracing with the current process parameters, extract key parameters such as electroplating current density and solution temperature as feature labels, and match the historical best baseline template through a deep learning model to update the baseline model feature library, and finally construct the baseline 3D model.

[0065] In this embodiment, the present invention utilizes the spectral dynamic matching technology of an adjustable spectral light source array to adjust the intensity ratio of blue light (450-495nm) to infrared (850-940nm) bands to an optimal value of 3:1 in real time according to the reflectivity of the copper foil material. Simultaneously, a ring polarizing filter (extinction ratio > 30dB) and diffuse blue light (scattering angle > 120°) are activated, reducing the specular reflection intensity of the metal surface by 92%, fundamentally solving the problem of optical measurement inaccuracies on highly reflective surfaces. A piezoelectric micro-displacement platform (resolution 0.1μm) drives a laser interferometer to generate an additive frequency of 10μm⁻¹ at the edge of a reference plate. A close-range positioning grating is used, and europium-based fluorescent marker arrays (50 μm in diameter) with a half-life of 8 seconds are sprayed into the non-detection area of ​​the sample through an aerosol nozzle (atomized particle size < 5 μm). This constructs a dual-source reference coordinate network with 2560 reference points, achieving sub-micron-level positioning accuracy under environmental vibration interference. Based on the deformation sequence captured by an industrial camera (frame rate 1000fps) and the height topology data obtained by a line structured light scanner (Z-axis resolution 0.2 μm), a dynamic mapping model between the flat grating coordinates and the fluorescent markers is established through a spatiotemporal correlation algorithm (fitting residual < 0.3 μm), eliminating measurement drift caused by micro-displacement of the sample.

[0066] In the 3D feature fusion stage, an improved optical flow algorithm (Lucas-Kanade variant) is used to align the micro-deformation trajectories of the sequence images, fuse height data to generate an initial point cloud (density 500 points / mm²), and use flat grating coordinates to calibrate the overall offset of the sample (compensation accuracy ±0.5μm); in the physical property decomposition stage, a material thermal expansion coefficient database is introduced (copper foil α=16.5×10⁻ 6 The temperature drift artifacts were removed by combining real-time temperature monitoring data (sampling rate 10Hz). At the same time, the reference surface error was eliminated by the plate deformation monitoring module (strain gauge accuracy 0.5με). Finally, the characteristic morphology features of the test sample were extracted (retention rate > 98%). The ideal plane equation of the reference plate was embedded in the bottom layer (least square fitting). The coordinate system transformation matrix based on quaternions was established in the middle layer (rotation accuracy 0.001°). The morphology vector containing curvature gradient features (128-dimensional feature descriptor) was integrated in the surface layer. The error tracing chain can be constructed to a three-dimensional reference model that is accurate to each feature point.

[0067] Furthermore, the electrolytic copper foil warpage continuity test method also includes the following steps:

[0068] Step S103: Perform three-dimensional panoramic scanning and multi-point scanning supplementation on the electrolytic copper foil sample, and perform spatiotemporal alignment and noise reduction processing on the data to obtain sample warping data.

[0069] In this embodiment of the invention, a three-dimensional panoramic scan of the electrolytic copper foil sample is performed to obtain the three-dimensional point cloud data of the electrolytic copper foil sample. The center position and four sides of the electrolytic copper foil sample are scanned to obtain supplementary scan data. The three-dimensional point cloud data and the supplementary scan data are then spatiotemporally aligned to generate aligned scan data. The aligned scan data is then denoised to identify and remove abnormal points such as water stains and transient dust, and the sample warping data is obtained.

[0070] Specifically, Figure 4 A flowchart illustrating the method for obtaining sample warpage data provided in an embodiment of the present invention is shown.

[0071] In a preferred embodiment of the present invention, the step of performing three-dimensional panoramic scanning and multi-point scanning supplementation on the electrolytic copper foil sample, and performing spatiotemporal alignment and noise reduction processing on the data to obtain sample warping data specifically includes the following steps:

[0072] Step S1031: Perform a three-dimensional panoramic scan on the electrolytic copper foil sample to obtain three-dimensional point cloud data;

[0073] Step S1032: Perform supplementary scanning of the center and four sides of the electrolytic copper foil sample to obtain supplementary scanning data;

[0074] Step S1033: Perform spatiotemporal alignment processing on the 3D point cloud data and the supplementary scan data to obtain aligned scan data;

[0075] Step S1034: Denoise the alignment scan data to obtain sample warping data.

[0076] In a preferred embodiment of the present invention, the process of performing spatiotemporal alignment processing on the 3D point cloud data and the supplementary scan data to obtain aligned scan data specifically includes the following steps:

[0077] Step S10341: Perform surface micro-protrusion feature recognition on the three-dimensional point cloud data to locate the core anchor point with spatial invariance. At the same time, identify the central cross mark point and edge feature inflection point in the supplementary scan data to generate a dual-source reference anchor point set.

[0078] Step S10342: Based on the spatiotemporal distribution of the dual-source reference anchor point set, analyze the timestamp sequence of the scanning probe's motion trajectory, and simultaneously invert the micro-displacement vector of the sample during the scanning process to generate a spatiotemporal displacement trajectory diagram.

[0079] Step S10343: Adaptive matching is performed on the spatiotemporal displacement trajectory map to obtain a multi-level optimized matching matrix; wherein, adaptive matching includes primary matching, secondary matching and tertiary matching; primary matching is to align the macroscopic position of the central cross marker point; secondary matching is to correct the regional offset of the inflection points of the four-sided features; tertiary matching is to achieve calibration by fusing micro-protrusion features;

[0080] Step S10344: Correct the multi-level optimized matching matrix, automatically identify and remove false feature points caused by dust adhesion, and repair local matching faults by obtaining the topological relationship of neighboring features. At the same time, verify the continuity of data boundaries and generate an anti-interference matching correction table.

[0081] Step S10345: Calculate the spatial rotation and translation parameters based on the anti-interference matching correction table, and map the supplementary scan data to the point cloud coordinate system in real time to achieve time window synchronization and obtain a spatiotemporally synchronized fusion dataset.

[0082] Step S10346: Randomly sample and verify the feature point alignment accuracy of the spatiotemporally synchronized fusion dataset, and check the data boundary overlap and matching degree. Generate a verification report with a confidence index. Then, based on the verified feature point alignment accuracy, data boundary overlap and matching degree, and the verification report with a confidence index, obtain quality verification alignment data.

[0083] Step S10347: Based on the feature point position deviation records in the quality verification alignment data, analyze the offset patterns of the central and edge regions; dynamically adjust the feature anchor point recognition sensitivity parameters according to the spatial distribution characteristics of the offset patterns; and optimize the acceleration compensation coefficient in the displacement trajectory reconstruction algorithm by combining the displacement trajectory features in the historical scan data to obtain the optimized parameters. The optimized parameters are then fed back to the anchor point extraction and trajectory reconstruction stages in real time to form a closed-loop self-optimization system, resulting in alignment scan data with continuous optimization capabilities. The alignment scan data with continuous optimization capabilities is then used as the final alignment scan data.

[0084] In this embodiment, the present invention employs multi-scale feature anchor point extraction technology: identifying surface micro-protrusion features (curvature threshold > 0.05 μm⁻¹, spatial invariance verification error < 0.8 μm) in three-dimensional point cloud data, and simultaneously locating the laser-engraved central cross mark (line width 10 μm) and edge feature inflection points (curvature change rate > 15%) in supplementary scanning data, constructing a dual-source reference anchor point set containing 512 spatial constraint points; based on the motion trajectory analysis module, analyzing the six-degree-of-freedom displacement of the scanning probe (linear encoder resolution 0.1 μm, gyroscope zero-bias stability 0.01° / h), inverting the micro-displacement vector of the sample (creep compensation model temperature coefficient -1.2% / ℃), and generating a spatiotemporal displacement trajectory map containing timestamps (synchronization accuracy ±0.1 ms);

[0085] A three-level adaptive matching engine was developed: primary matching aligns the center cross markers using least-squares fitting (position error < 5 μm); secondary matching uses Thin Plate Spline (TPS) function to correct the offset of the inflection point region of the four-sided features (residual < 2 μm); and tertiary matching uses an improved ICP algorithm (point-to-plane error function) to fuse micro-protrusion features (registration accuracy 0.7 μm). The correction module integrates a convolutional neural network (CNN) pseudo-feature recognition model (training samples contain 8 types of dust interference features), automatically removes pseudo-feature points (recognition rate > 95%), and reconstructs neighboring features through Delaunay triangulation. Topological relationships are established to repair local matching faults (boundary continuity error < 0.4 μm); during the spatial transformation stage, the Kabsch algorithm is used to calculate rotation and translation parameters (root mean square error < 1.2 μm), and the supplementary scan data is mapped to the point cloud coordinate system in real time; the closed-loop self-optimization system establishes an offset pattern library based on 300 sets of historical scan data, dynamically adjusts the feature anchor point recognition sensitivity parameters (adaptive range ± 40%), optimizes the acceleration compensation coefficient of trajectory reconstruction (Kalman filter Q matrix is ​​dynamically adjusted), and improves the alignment accuracy with the number of iterations (accuracy reaches 0.3 μm after the 100th iteration).

[0086] Furthermore, the electrolytic copper foil warpage continuity test method also includes the following steps:

[0087] Step S104: Based on the warping data of the sample, reconstruct a three-dimensional surface model, perform benchmark comparison, and calculate multiple key warping indices.

[0088] In this embodiment of the invention, a three-dimensional surface model of the electrolytic copper foil sample is reconstructed based on the sample warpage data. The three-dimensional surface model is then compared with a reference three-dimensional model and comparative identification is performed. The maximum warpage height, average warpage angle, warpage distribution curvature, and warpage asymmetry index of the electrolytic copper foil sample are calculated, thereby obtaining multiple key warpage indicators of the electrolytic copper foil sample.

[0089] In a preferred embodiment of the present invention, the step of reconstructing a three-dimensional surface model based on the warping data of the sample, performing benchmark comparison, and calculating multiple key warping indicators specifically includes the following steps:

[0090] Step S1041: Perform copper foil surface characteristic identification on the discrete point cloud in the sample warping data, and automatically divide the high-sensitivity area (corresponding to micro-wrinkles) and low-sensitivity area (corresponding to smooth surface) according to the rate of change of regional curvature to generate a feature sensitivity partition map.

[0091] Step S1042: Based on the feature sensitivity partitioning map, a mutation feature preservation mechanism is enabled in the high-sensitivity area, edge sharpness is maintained by the spatial vector constraint of the neighboring point cloud, and curvature smoothing transfer processing is performed in the low-sensitivity area to ensure the continuous transition of the normal vectors of the adjacent point clouds, thus obtaining a feature-fidelity point cloud set.

[0092] Step S1043: Perform cross-scale fusion on the feature-fidelity point cloud to generate a cross-scale enhanced point cloud topology; wherein, cross-scale fusion includes macro-fusion and micro-fusion; macro-fusion enhances the spatial weight of the key points of the contour in the four-sided region; micro-fusion improves the topological correlation of the texture feature points in the central region.

[0093] Step S1044: Based on the spatial gradient distribution of cross-scale enhanced point cloud topology, an initial surface patch is generated with the central feature point as the growth origin, and the surface mesh is dynamically expanded along the curvature change direction. The curvature continuity threshold of the mesh nodes is simultaneously verified in real time to obtain the dynamically expanded surface framework.

[0094] Step S1045: Perform iterative correction on the dynamically extended surface framework, detect local curvature abrupt change regions and mark abnormal nodes, reconstruct a smooth transition surface based on the spatial distribution of neighboring point clouds, simultaneously verify the differential continuity index of the corrected surface, and generate a self-optimizing surface network.

[0095] Step S1046: Associate the self-optimizing surface network with the current process parameters, extract the electroplating current density feature value to adjust the surface fitting sensitivity, optimize the surface smoothing threshold according to the solution temperature parameter, and simultaneously match the historical best reconstruction parameters to achieve dynamic calibration and obtain a preliminary three-dimensional surface model.

[0096] Step S1047: Randomly extract profile lines from the preliminary three-dimensional surface model to verify the curvature continuity, compare the feature fidelity with the original point cloud data, and generate a quality traceability map with spatial error distribution. Finally, a traceable benchmark three-dimensional surface model is constructed and used as the final reconstructed three-dimensional surface model.

[0097] Step S1048: Compare and identify the three-dimensional surface model with the reference three-dimensional model, and calculate the maximum warpage height, average warpage angle, warpage distribution curvature and warpage asymmetry index of the electrolytic copper foil sample.

[0098] In this embodiment, the present invention develops a copper foil surface curvature sensitivity analysis algorithm: based on the regional curvature change rate (threshold Kc=0.03μm⁻¹), it automatically divides high-sensitivity areas (micro-wrinkle areas, accounting for 12±3%) and low-sensitivity areas (smooth areas), generating a feature sensitivity partition map (grid resolution 50μm); in the high-sensitivity areas, a mutation feature retention mechanism is adopted, and edge sharpness is maintained (curvature retention rate > 90%) by constructing a spatial vector constraint field of neighboring point clouds (constraint force coefficient λ=0.85); in the low-sensitivity areas, curvature smoothing transfer processing is implemented, and bicubic B-spline basis functions are used to ensure continuous transition of normal vectors (normal vector angle change < 1°); an innovative cross-scale fusion architecture is used: macro-fusion assigns 3 times the spatial weight to the key points of the four-sided contour (based on the contribution of contour curvature), and micro-fusion improves the topological correlation of texture feature points in the central region through graph neural networks (GNN) (the number of adjacent nodes increases to 32).

[0099] Based on spatial gradient distribution (Sobel operator enhancement), NURBS surface patches are generated with the central feature point as the growth origin. The Catmull-Clark subdivision surface mesh is dynamically expanded along the principal curvature direction (κ1, κ2), and the curvature continuity threshold of the mesh nodes is verified in real time (Δκ < 0.005 μm⁻¹). An iterative correction module integrates an anomaly node detection model (Z-score > 3 marks anomalies), uses radial basis function (RBF) kernels to reconstruct a smooth transition surface, and generates a self-optimizing surface network through differential continuity verification (C² continuity compliance rate > 98%). In the dynamic coupling stage of process parameters, the electroplating current density is extracted. The sensitivity of surface fitting is adjusted by using eigenvalues ​​(range 10-50 ASD) (linear relationship coefficient β=0.12 / ASD). The surface smoothing threshold is optimized based on the solution temperature parameter (20-50℃) (exponential relationship γ=0.8^((T-25) / 5)). Dynamic calibration is achieved by synchronously matching the historical best reconstruction parameter library (containing 200 process combinations). Finally, the curvature continuity is verified by randomly selecting 100 profile lines through Monte Carlo sampling (compliance rate >95%), the feature fidelity is calculated (>97%), and a heat map with spatial error distribution (RGB encoded error value) is generated to construct a traceable benchmark 3D surface model.

[0100] In a preferred embodiment of the present invention, the step of comparing and identifying the three-dimensional surface model with the reference three-dimensional model, and calculating the maximum warpage height, average warpage angle, warpage distribution curvature, and warpage asymmetry index of the electrolytic copper foil sample specifically includes the following steps:

[0101] Step S10481: Extract the set of micro-wrinkle feature points in the three-dimensional surface model and the corresponding reference point set in the benchmark three-dimensional model, and generate a feature point sensitivity weight distribution map based on the material characteristics of electrolytic copper foil to obtain a feature weight mapping table.

[0102] Step S10482: Based on the plate deformation data recorded in the reference three-dimensional model, perform real-time deformation correction on the corresponding area of ​​the three-dimensional surface model, and at the same time use the temperature coefficient in the sample process parameters to compensate for the thermal expansion deviation, so as to obtain the compensated three-dimensional surface model.

[0103] Step S10483: Align the compensated 3D surface model with the origin of the central coordinate system of the reference 3D model, match the feature points of the four-sided contour, and simultaneously perform local calibration on the micro-wrinkle feature points to obtain the spatial registration relationship matrix.

[0104] Step S10484: Based on the electroplating current density parameter in the sample identity information, separate the inherent deformation component caused by uneven current distribution, extract the pure material warping features, and generate a warping dataset with process decoupling.

[0105] Step S10485: Establish a dynamic comparison grid based on the spatial registration relationship matrix, calculate the angle between the normal vectors at each grid node and mark the abnormal deformation area to obtain a dynamic deformation comparison map;

[0106] Step S10486: Based on the dynamic deformation comparison map, traverse all grid nodes to extract the maximum normal vector angle as the warpage height index; based on the warpage dataset of process decoupling, calculate the ratio of the area marked as abnormal deformation to the total area as the distribution curvature index; based on the warpage dataset of process decoupling, calculate the standard deviation of the deformation in its four quadrants as the warpage asymmetry index; combine the warpage height index, the distribution curvature index, and the warpage asymmetry index to generate a multidimensional warpage index set;

[0107] Step S10487: Associate each indicator in the multidimensional warp index set with its corresponding spatial coordinate region and process parameters to finally obtain a traceable set of key warp indicators.

[0108] Step S10488: Based on the traceable set of key warping indicators, locate the maximum deformation region corresponding to the peak of the normal vector angle in the spatial coordinate region, and extract the peak value as the maximum warping height of the electrolytic copper foil sample; calculate the weighted average of the normal vector angles of all grid nodes as the average warping angle of the sample; statistically analyze the weight ratio of the area of ​​the abnormal deformation region to the total area to generate the warping distribution curvature; calculate the standard deviation of the deformation in the four quadrants as the warping asymmetry index.

[0109] In this embodiment, the present invention establishes a material property sensitivity weighting model: based on the electrolytic copper foil grain orientation database (EBSD data), a feature point sensitivity weight distribution map (weighting coefficient ω=0.2~1.5) is generated to quantify the warping contribution of different regions; in the deformation correction stage: based on the plate deformation data recorded by the benchmark three-dimensional model (strain resolution 0.1με), finite element analysis (FEA) is used to correct the corresponding region of the three-dimensional surface model in real time (stiffness matrix update frequency 10Hz), and simultaneously based on the temperature coefficient (α=16.5×10⁻) in the sample process parameters. 6 / ℃) to compensate for thermal expansion deviation; spatial registration is achieved by quaternion coordinate system transformation to align the center origin (translation residual <0.5μm), and the four-sided contour feature points are matched using the improved Procrustes algorithm (rotation error <0.01°), and the micro-wrinkle feature points are locally calibrated using the non-rigid ICP algorithm (local deformation field modeling); process decoupling stage: based on the electroplating current density distribution map (scanning accuracy 1mm) 2 The inherent deformation components caused by uneven current are separated by backpropagation neural network (BPNN) (decoupling rate > 88%), and the intrinsic warping features of the pure material are extracted.

[0110] A dynamic comparison mesh system (mesh size 1mm×1mm) is constructed. At each node, the angle between the normal vectors is calculated as θ = arccos(n1·n2) (accuracy 0.01°), and abnormal deformation regions (θ > 0.5°) are marked. Multidimensional index calculation: The maximum angle between the normal vectors θ is extracted. max The area ratio of the abnormal region (threshold θ > 0.3°) is calculated as the warp height (range 0-5mm) and the distribution curvature k is used as the distribution curvature. d Analysis of the standard deviation σ of deformation in the four quadrants quad (Sampling points 256 / quadrant) Generate the asymmetric index ASI=σ quad / θ max Finally, a spatial-process correlation matrix was established, mapping each indicator to the corresponding coordinate region (positioning accuracy ±0.2mm) and associating it with 32 process parameters such as electroplating current density and temperature, forming a set of key warping indicators that are traceable throughout the entire process.

[0111] Furthermore, the electrolytic copper foil warpage continuity test method also includes the following steps:

[0112] Step S105: Perform qualification judgment and trend analysis on multiple key warpage indicators, generate a warpage test report, and perform feedback control for continuous testing.

[0113] In this embodiment of the invention, test threshold data is extracted from the process parameters of the sample identity information. Then, multiple warpage key indicators are compared with the corresponding thresholds in the test threshold data to determine whether the electrolytic copper foil sample is qualified. The judgment result is recorded, and the trend analysis of historical judgment results is performed. Then, based on the three-dimensional surface model and multiple warpage key indicators, a three-dimensional surface map and a warpage distribution curve are generated. By filling in the judgment result, the three-dimensional surface map and the warpage distribution curve into the report, a warpage test report of the electrolytic copper foil sample is generated. If the judgment result is unqualified or has a trend of unqualification, the process feedback control of continuous testing is triggered, such as reducing the electroplating current density, adjusting the tension or temperature, etc.

[0114] Specifically, Figure 5 A flowchart illustrating the generation of a warpage test report in the method provided by an embodiment of the present invention is shown.

[0115] In a preferred embodiment of the present invention, the step of judging the pass / fail status and trend analysis of multiple key warpage indicators, generating a warpage test report, and performing feedback control for continuous testing specifically includes the following steps:

[0116] Step S1051: Extract test threshold data from the process parameters of the sample identity information;

[0117] Step S1052: Compare multiple warpage key indicators with the test threshold data to determine whether the electrolytic copper foil sample is qualified, perform trend analysis, and record the judgment results;

[0118] Step S1053: Generate a three-dimensional surface map and a warping distribution curve based on the three-dimensional surface model and multiple warping key indicators;

[0119] Step S1054: Based on the judgment result, the three-dimensional surface plot, and the warpage distribution curve, generate a warpage test report;

[0120] Step S1055: When the judgment result is unqualified or has a trend of unqualification, the process feedback control of continuous testing is triggered.

[0121] Furthermore, Figure 6 An application architecture diagram of the system provided in an embodiment of the present invention is shown.

[0122] In another preferred embodiment of the present invention, the electrolytic copper foil warpage continuity testing system includes:

[0123] The copper foil sample processing unit 101 is used to periodically cut electrolytic copper foil samples, record sample identification information, and transport them to the warpage test channel.

[0124] In this embodiment of the invention, before the formal production of electrolytic copper foil, the copper foil sample processing unit 101 determines the copper foil sampling cycle and copper foil sampling parameters. According to the copper foil sampling cycle and copper foil sampling parameters, electrolytic copper foil samples are periodically cut on the electrolytic copper foil production line. At the same time as the electrolytic copper foil samples are cut, a copper foil sample ID is generated, and the production batch, production time and process parameters corresponding to the electrolytic copper foil sample are obtained. By comprehensively organizing the copper foil sample ID, production batch, production time and process parameters, the sample identity information is recorded, and then the electrolytic copper foil sample is transported to the warpage test channel.

[0125] Specifically, Figure 7 A structural block diagram of the copper foil sample processing unit 101 in the system provided in an embodiment of the present invention is shown.

[0126] In a preferred embodiment of the present invention, the copper foil sample processing unit 101 specifically includes:

[0127] The period parameter determination module 1011 is used to determine the copper foil sampling period and copper foil sampling parameters;

[0128] The sample cutting module 1012 is used to periodically cut electrolytic copper foil samples on the electrolytic copper foil production line according to the copper foil sampling cycle and the copper foil sampling parameters.

[0129] ID generation module 1013 is used to generate copper foil sample ID;

[0130] Information acquisition module 1014 is used to acquire the production batch, production time and process parameters corresponding to the electrolytic copper foil sample;

[0131] The identity information recording module 1015 is used to record the sample identity information by comprehensively considering the copper foil sample ID, the production batch, the production time, and the process parameters.

[0132] The sample delivery module 1016 is used to deliver the electrolytic copper foil sample to the warpage test channel.

[0133] Furthermore, the continuous testing system for electrolytic copper foil warpage also includes:

[0134] The test environment adjustment unit 102 is used to adjust the test environment according to the sample identity information, and to perform flat positioning and benchmark calibration on the electrolytic copper foil sample.

[0135] Specifically, Figure 8 A structural block diagram of the test environment adjustment unit 102 in the system provided in an embodiment of the present invention is shown.

[0136] In a preferred embodiment of the present invention, the test environment adjustment unit 102 specifically includes:

[0137] The environmental data extraction module 1021 is used to extract test environment data from the process parameters of the sample identity information and determine the reference plate.

[0138] The environment adjustment module 1022 is used to adjust the test environment according to the test environment data;

[0139] The leveling and positioning module 1023 is used to level and position the electrolytic copper foil sample on the reference plate.

[0140] The 3D model construction module 1024 is used to perform optical benchmark calibration on the flat and positioned electrolytic copper foil sample and construct a benchmark 3D model.

[0141] Furthermore, the continuous testing system for electrolytic copper foil warpage also includes:

[0142] The three-dimensional scanning processing unit 103 is used to perform three-dimensional panoramic scanning and multi-point scanning supplementation on the electrolytic copper foil sample, and to perform spatiotemporal alignment and noise reduction processing on the data to obtain sample warping data.

[0143] The model benchmark comparison unit 104 is used to reconstruct a three-dimensional surface model based on the warping data of the sample, perform benchmark comparison, and calculate multiple key warping indicators.

[0144] In this embodiment of the invention, the model benchmark comparison unit 104 reconstructs the three-dimensional surface model of the electrolytic copper foil sample based on the sample warpage data, and then compares and identifies the three-dimensional surface model with the benchmark three-dimensional model to calculate the maximum warpage height, average warpage angle, warpage distribution curvature and warpage asymmetry index of the electrolytic copper foil sample, thereby obtaining multiple key warpage indicators of the electrolytic copper foil sample.

[0145] The test report generation unit 105 is used to perform qualification judgment and trend analysis on multiple warpage key indicators, generate a warpage test report, and perform feedback control for continuous testing.

[0146] Specifically, Figure 9 The diagram shows the structural block diagram of the test report generation unit 105 in the system provided by an embodiment of the present invention.

[0147] In a preferred embodiment of the present invention, the test report generation unit 105 specifically includes:

[0148] The threshold data extraction module 1051 is used to extract test threshold data from the process parameters of the sample identity information;

[0149] The comparison and analysis module 1052 is used to compare multiple warpage key indicators with the test threshold data, determine whether the electrolytic copper foil sample is qualified, perform trend analysis, and record the judgment results.

[0150] The model index processing module 1053 is used to generate a three-dimensional surface plot and a warping distribution curve based on the three-dimensional surface model and multiple warping key indicators.

[0151] The test report generation module 1054 is used to generate a warpage test report by integrating the judgment result, the three-dimensional surface plot and the warpage distribution curve.

[0152] The process feedback control module 1055 is used to trigger the process feedback control of continuous testing when the judgment result is unqualified or has a trend of unqualification.

[0153] It should be understood that although the steps in the flowcharts of the various embodiments of the present invention are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the various embodiments may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least a portion of the sub-steps or stages of other steps.

[0154] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0155] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0156] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

[0157] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An electrolytic copper foil warpage continuous test method characterized by, The method specifically includes the following steps: Electrolytic copper foil samples are periodically cut, sample identification information is recorded, and the samples are transported to the warpage test channel. Based on the sample identification information, the test environment is adjusted, and the electrolytic copper foil sample is flattened, positioned, and calibrated. The electrolytic copper foil sample was subjected to three-dimensional panoramic scanning and multi-point scanning supplementation, and the data was spatiotemporally aligned and denoised to obtain sample warping data. Based on the warping data of the sample, a three-dimensional surface model is reconstructed, and a benchmark comparison is performed to calculate multiple key warping indicators. The pass / fail judgment and trend analysis of multiple key warpage indicators are performed, a warpage test report is generated, and feedback control for continuous testing is implemented. The process of performing three-dimensional panoramic scanning and multi-point scanning supplementation on the electrolytic copper foil sample, and performing spatiotemporal alignment and noise reduction on the data to obtain sample warping data specifically includes the following steps: The electrolytic copper foil sample was subjected to a three-dimensional panoramic scan to obtain three-dimensional point cloud data; The electrolytic copper foil sample is scanned at its center and four sides to obtain supplementary scan data; The three-dimensional point cloud data and the supplementary scan data are spatiotemporally aligned to obtain aligned scan data. The alignment scan data is denoised to obtain sample warping data; The process of performing spatiotemporal alignment processing on the 3D point cloud data and the supplementary scan data to obtain aligned scan data specifically includes the following steps: Surface micro-protrusion feature recognition is performed on 3D point cloud data to locate the core anchor point with spatial invariance. At the same time, the central cross mark point and edge feature inflection point are identified in the supplementary scan data to generate a dual-source reference anchor point set. Based on the spatiotemporal distribution of the dual-source reference anchor point set, the timestamp sequence of the scanning probe motion trajectory is analyzed, and the micro-displacement vector of the sample during the scanning process is inverted to generate a spatiotemporal displacement trajectory map. Adaptive matching is performed on the spatiotemporal displacement trajectory map to obtain a multi-level optimized matching matrix. The adaptive matching includes primary matching, secondary matching, and tertiary matching. Primary matching aligns the macroscopic position of the central cross marker point; secondary matching corrects the regional offset of the inflection points of the four-sided features; and tertiary matching achieves calibration by fusing micro-protrusion features. The multi-level optimized matching matrix is ​​corrected, and false feature points caused by dust adhesion are automatically identified and removed. The local matching faults are repaired by obtaining the topological relationship of neighboring features. At the same time, the continuity of data boundaries is verified and an anti-interference matching correction table is generated. Spatial rotation and translation parameters are calculated based on the anti-interference matching correction table, and the supplementary scan data is mapped to the point cloud coordinate system in real time to achieve time window synchronization and obtain a spatiotemporally synchronized fusion dataset. Random sampling is performed on the spatiotemporally synchronized fusion dataset to verify the feature point alignment accuracy, while checking the data boundary overlap and matching degree. A verification report with a confidence index is generated, and then the quality verification alignment data is obtained based on the verified feature point alignment accuracy, data boundary overlap and matching degree, and the verification report with a confidence index. Based on the feature point position deviation records in the quality verification alignment data, the offset patterns of the central and edge regions are analyzed; the feature anchor point recognition sensitivity parameters are dynamically adjusted according to the spatial distribution characteristics of the offset patterns; and the acceleration compensation coefficient in the displacement trajectory reconstruction algorithm is optimized by combining the displacement trajectory features in the historical scan data to obtain the optimized parameters. The optimized parameters are fed back to the anchor point extraction and trajectory reconstruction stages in real time, forming a closed-loop self-optimizing system to obtain aligned scan data with continuous optimization capabilities. This aligned scan data with continuous optimization capabilities is used as the final aligned scan data.

2. The electrolytic copper foil warp continuous test method according to claim 1, characterized by, The process of periodically cutting electrolytic copper foil samples, recording sample identification information, and transporting them to the warpage test channel specifically includes the following steps: Determine the copper foil sampling cycle and copper foil sampling parameters; According to the copper foil sampling cycle and the copper foil sampling parameters, electrolytic copper foil samples are periodically cut on the electrolytic copper foil production line. Generate copper foil sample ID; Obtain the production batch, production time, and process parameters corresponding to the electrolytic copper foil sample; Based on the copper foil sample ID, the production batch, the production time, and the process parameters, record the sample identification information; The electrolytic copper foil sample is transported to the warpage test channel.

3. The electrolytic copper foil warp continuous test method according to claim 2, characterized by, The process of adjusting the testing environment based on the sample identification information and performing flattening, positioning, and benchmark calibration on the electrolytic copper foil sample specifically includes the following steps: From the process parameters of the sample identification information, test environment data is extracted to determine the reference plate. Adjust the test environment according to the test environment data described above; The electrolytic copper foil sample is flattened and positioned on the reference plate. Optical benchmark calibration was performed on the flat and positioned electrolytic copper foil sample to construct a benchmark three-dimensional model.

4. The method for continuous testing of electrolytic copper foil warpage according to claim 3, characterized in that, The process of optically calibrating the flat and positioned electrolytic copper foil sample and constructing a reference three-dimensional model specifically includes the following steps: An array of tunable spectral light sources is deployed on the flat and positioned surface of an electrolytic copper foil sample. The intensity ratio of blue light band and infrared band is matched according to the reflective characteristics of the electrolytic copper foil sample material. The ring polarization filter and diffuse blue light in the light source are activated simultaneously to suppress glare on the metal surface and generate an anti-interference composite lighting field. In an anti-interference composite illumination field, a piezoelectric micro-displacement driven laser interferometer is used to generate a dense positioning grating on the edge of a reference plate, and an aerosol nozzle is controlled to spray a transient fluorescent marker array in the non-detection area of ​​the sample to form a dual-source reference coordinate network. Based on the spatiotemporal correlation characteristics of the dual-source reference coordinate network, an industrial camera is driven to capture surface deformation sequences, and a line structured light scanner is activated to acquire height topology data. The spatial mapping relationship between the flat grating and the sample marker points is recorded simultaneously to obtain the original morphology dataset. Three-dimensional feature fusion is performed on the original morphology dataset to generate a fused feature model with spatial constraints. The three-dimensional feature fusion on the original morphology dataset includes: aligning the micro-deformation trajectories of the sequence images using an optical flow algorithm; constructing an initial surface point cloud by fusing height topology data; and calibrating the overall offset of the sample using flat grating coordinates. Physical property decomposition is performed on the spatially constrained fusion feature model to obtain the artifact-free intrinsic morphology point cloud. The physical property decomposition of the spatially constrained fusion feature model includes: removing environmental temperature drift artifacts based on the material's thermal expansion coefficient; acquiring plate deformation monitoring data and eliminating reference surface errors through the plate deformation monitoring data; and preserving the inherent micromorphological features of the sample. The artifact-free intrinsic morphology point cloud is layered to generate a baseline 3D model for error tracing. The layering of the artifact-free intrinsic morphology point cloud includes: embedding the ideal plane equation of the reference plate in the bottom layer; establishing the sample coordinate system transformation matrix in the middle layer; and integrating the micro-morphology feature vector in the surface layer. The baseline 3D model for error tracing is associated with the current process parameters. Key parameters such as electroplating current density and solution temperature are extracted as feature labels. The baseline model feature library is updated by matching the historical best baseline template through a deep learning model, and finally the baseline 3D model is constructed.

5. The electrolytic copper foil warp continuous test method according to claim 4, characterized by, The process of reconstructing a three-dimensional surface model based on the warpage data of the sample, performing benchmark comparisons, and calculating multiple key warpage indicators specifically includes the following steps: The surface characteristics of copper foil are identified by performing discrete point cloud analysis on the sample warping data. Based on the rate of change of regional curvature, high-sensitivity and low-sensitivity areas are automatically divided to generate a feature sensitivity partition map. Based on the feature sensitivity partitioning map, a mutation feature preservation mechanism is enabled in the high-sensitivity area, edge sharpness is maintained by the spatial vector constraint of the neighboring point cloud, and curvature smoothing transfer processing is performed in the low-sensitivity area to ensure the continuous transition of the normal vectors of the adjacent point clouds, thus obtaining a feature-fidelity point cloud set. Cross-scale fusion is performed on the feature-fidelity point cloud to generate a cross-scale enhanced point cloud topology. The cross-scale fusion includes macro-fusion and micro-fusion. Macro-fusion enhances the spatial weight of key points in the contour of the four-sided region. Micro-fusion improves the topological correlation of texture feature points in the central region. Based on the spatial gradient distribution of cross-scale enhanced point cloud topology, an initial surface patch is generated with the central feature point as the growth origin, and the surface mesh is dynamically expanded along the curvature change direction. The curvature continuity threshold of the mesh nodes is simultaneously verified in real time to obtain a dynamically expanded surface framework. Iterative correction is performed on the dynamically extended surface framework, local curvature abrupt regions are detected and abnormal nodes are marked, and a smooth transition surface is reconstructed based on the spatial distribution of neighboring point clouds. The differential continuity index of the corrected surface is verified simultaneously, and a self-optimizing surface network is generated. By associating the self-optimizing surface network with the current process parameters, extracting the electroplating current density feature value to adjust the surface fitting sensitivity, and optimizing the surface smoothing threshold according to the solution temperature parameter, the dynamic calibration is achieved by synchronously matching the historical best reconstruction parameters, and a preliminary three-dimensional surface model is obtained. The curvature continuity of the preliminary 3D surface model is verified by randomly selecting cross-sections, and the feature fidelity is calculated by comparing it with the original point cloud data. A quality traceability map with spatial error distribution is generated, and finally a traceable benchmark 3D surface model is constructed. The traceable benchmark 3D surface model is used as the final reconstructed 3D surface model. The three-dimensional surface model is compared and identified with the reference three-dimensional model, and the maximum warpage height, average warpage angle, warpage distribution curvature, and warpage asymmetry index of the electrolytic copper foil sample are calculated.

6. The electrolytic copper foil warp continuous test method according to claim 5, characterized by, The step of comparing and identifying the three-dimensional surface model with the reference three-dimensional model, and calculating the maximum warpage height, average warpage angle, warpage distribution curvature, and warpage asymmetry index of the electrolytic copper foil sample specifically includes the following steps: Extract the set of micro-wrinkle feature points from the 3D surface model and the corresponding reference point set from the baseline 3D model, and generate a feature point sensitivity weight distribution map based on the material properties of electrolytic copper foil to obtain a feature weight mapping table. Based on the plate deformation data recorded in the benchmark three-dimensional model, the corresponding area of ​​the three-dimensional surface model is deformed in real time. At the same time, the temperature coefficient in the sample process parameters is used to compensate for the thermal expansion deviation, and the compensated three-dimensional surface model is obtained. The compensated 3D surface model is aligned with the origin of the central coordinate system with the reference 3D model, and the feature points of the four sides are matched. Simultaneously, the micro-wrinkle feature points are locally calibrated to obtain the spatial registration relationship matrix. Based on the electroplating current density parameters in the sample identification information, the inherent deformation components caused by uneven current distribution are separated, the pure material warping features are extracted, and a warping dataset with process decoupling is generated. A dynamic comparison grid is established based on the spatial registration relationship matrix. The angle between the normal vectors is calculated at each grid node and the abnormal deformation area is marked to obtain a dynamic deformation comparison map. Based on the dynamic deformation comparison map, the maximum normal vector angle is extracted from all grid nodes as a warping height index; based on the warping dataset of the process decoupling, the ratio of the area of ​​the region marked as abnormal deformation to the total area is calculated as a distribution curvature index. Based on the warp dataset of the process decoupling, the standard deviation of the deformation in the four quadrants of the warp dataset of the process decoupling is calculated as the warp asymmetry index; the warp height index, the distribution curvature index and the warp asymmetry index are aggregated to generate a multidimensional warp index set; By associating each multidimensional warp index with its corresponding spatial coordinate region and process parameters, a traceable set of key warp indices is finally obtained. Based on a traceable set of key warpage indicators, the maximum deformation region corresponding to the peak of the normal vector angle in the spatial coordinate region is located, and this peak is extracted as the maximum warpage height of the electrolytic copper foil sample. The weighted average of the normal vector angles of all grid nodes is calculated as the average warpage angle of the sample. The area of ​​the abnormal deformation region and the weight ratio of the total area are statistically analyzed to generate the warpage distribution curvature. The standard deviation of the deformation in the four quadrants is calculated as the warpage asymmetry index.

7. The electrolytic copper foil warp continuous test method according to claim 6, characterized by, The process of determining the pass / fail status and trend analysis of multiple key warpage indicators, generating a warpage test report, and implementing feedback control for continuous testing specifically includes the following steps: Extract test threshold data from the process parameters of the sample identification information; The multiple warpage key indicators are compared with the test threshold data to determine whether the electrolytic copper foil sample is qualified, and trend analysis is performed and the judgment results are recorded. Based on the three-dimensional surface model and multiple key warping indicators, a three-dimensional surface plot and a warping distribution curve are generated. Based on the judgment results, the three-dimensional surface plot, and the warpage distribution curve, a warpage test report is generated. When the judgment result is unqualified or has a trend of unqualification, the process feedback control of continuous testing is triggered.

8. An electrolytic copper foil warpage continuous test system, characterized by, The system employs the electrolytic copper foil warpage continuity testing method as described in any one of claims 1 to 7, and the system comprises: The copper foil sample processing unit is used to periodically cut electrolytic copper foil samples, record sample identification information, and transport them to the warpage test channel. The test environment adjustment unit is used to adjust the test environment according to the sample identification information, and to perform flat positioning and benchmark calibration on the electrolytic copper foil sample. The three-dimensional scanning processing unit is used to perform three-dimensional panoramic scanning and multi-point scanning supplementation on the electrolytic copper foil sample, and to perform spatiotemporal alignment and noise reduction processing on the data to obtain sample warping data. The model benchmark comparison unit is used to reconstruct a three-dimensional surface model based on the warping data of the sample, perform benchmark comparison, and calculate multiple key warping indicators. The test report generation unit is used to perform qualification judgment and trend analysis on multiple key warpage indicators, generate a warpage test report, and perform feedback control for continuous testing.

9. The electrolytic copper foil warp continuous test system of claim 8, wherein, The copper foil sample processing unit specifically includes: The period parameter determination module is used to determine the copper foil sampling period and copper foil sampling parameters; The sample cutting module is used to periodically cut electrolytic copper foil samples on the electrolytic copper foil production line according to the copper foil sampling cycle and the copper foil sampling parameters. The ID generation module is used to generate copper foil sample IDs; The information acquisition module is used to acquire the production batch, production time and process parameters corresponding to the electrolytic copper foil sample; The identity information recording module is used to record the sample identity information by combining the copper foil sample ID, the production batch, the production time, and the process parameters. The sample delivery module is used to deliver the electrolytic copper foil sample to the warpage test channel.

Images