CN-122016935-A - Intelligent detection method and system for defects of ultrathin electrolytic copper foil facing new energy field

CN122016935ACN 122016935 ACN122016935 ACN 122016935ACN-122016935-A

Abstract

The invention discloses an intelligent detection method and system for defects of an ultrathin electrolytic copper foil facing the new energy field, which belong to the technical field of intelligent detection and comprise the following steps of continuously collecting an original image sequence with fixed line frequency, and carrying out visual detection on a preprocessed image to judge and obtain a suspected defect region set; generating an electrical measurement strategy for the electrical verification task to be triggered, triggering local electrical measurement according to the generated electrical measurement strategy, acquiring local measurement data of the suspected region, and judging the suspected result of the suspected defect region. The invention solves the problem of low defect type detection accuracy caused by a single detection means in the high-speed production process in the prior art by constructing a collaborative detection flow taking visual detection as an entrance and electrical detection as a pertinence verification means.

Inventors

ZHANG REN
LIAO PINGYUAN
YANG JIANWEN
YANG YUPING
HE YIMAO
ZHU YUSEN

Assignees

广东嘉元科技股份有限公司

Dates

Publication Date: 20260512
Application Date: 20260127

Claims (10)

1. The intelligent detection method for the defects of the ultrathin electrolytic copper foil facing the new energy field is characterized by comprising the following steps of: S1, continuously acquiring an original image sequence with fixed line frequency when an ultrathin electrolytic copper foil passes through an industrial camera field of view, preprocessing an original image, and performing visual detection on the preprocessed image to judge to obtain a suspected defect region set; S2, performing space-time alignment on the suspected defect area set obtained by visual detection, generating an electrical verification task queue according to the coil running process, and simultaneously executing scheduling on the task queue; s3, generating an electrical measurement strategy aiming at an electrical verification task to be triggered, wherein the electrical measurement strategy comprises an electrical measurement mode, a set electrical scanning path and points; And S4, triggering local electrical measurement according to the generated electrical measurement strategy by using an electrical detection probe, acquiring local measurement data of a suspected region, and judging a suspected result of the suspected defect region where the electrical measurement is completed.
2. The intelligent detection method for defects of the ultrathin electrolytic copper foil facing the new energy field, which is characterized by comprising the steps of preprocessing an original image and visually detecting the preprocessed image, wherein the method comprises the following steps of: performing illumination and background compensation on the original image based on the set reference image to inhibit fixed mode interference; performing adaptive enhancement of limited contrast on the compensated image to highlight weak texture features on the surface of the copper foil; Performing spatial distortion correction on the image, and uniformly mapping pixel coordinates to a physical coordinate system of the coiled material so as to provide a consistent spatial reference for cross-frame correlation and defect position tracking; Constructing a trans-scale local texture disturbance response set for representing copper foil surface abnormality based on the preprocessed image, wherein the local texture disturbance response set is at least used for representing local texture mutation, direction continuity unbalance and local statistical structure deviation; And performing unified scale alignment and difference enhancement processing on the cross-scale local texture disturbance response set, inhibiting interference of a background area on abnormal response, and fusing the processed response results to generate a saliency distribution map, wherein the saliency distribution map is used for representing the abnormal saliency degree of an image pixel or a local area relative to the neighborhood of the image pixel or the local area.
3. The intelligent detection method for defects of the ultrathin electrolytic copper foil facing the new energy field, which is characterized in that the judging method is used for obtaining a suspected defect area set, and comprises the following specific steps: Extracting a connected region with concentrated saliency responses from the saliency distribution map according to a saliency criterion for distinguishing abnormal responses from background responses and by combining with a spatial association relation between adjacent pixels or local regions, wherein the connected region is used for organizing pixel-level abnormal responses into candidate regions with spatial continuity so as to serve as a basic unit for generating suspected defect regions; calculating region features for each connected region to form a region-level multi-feature description vector, wherein the region features are used for representing region morphology, texture disturbance intensity and response consistency in a preset frame window; Generating visual confidence coefficient based on the multi-feature description vector, and carrying out category identification on the connected region according to components of the representation form and the texture of the multi-feature description vector, wherein the category identification at least comprises one of punctiform abnormality, linear deformation, planar disturbance or roughness abnormality; And forming a suspected defect region set containing the space characterization information, the visual confidence coefficient and the category identification of the suspected defect region, wherein the space characterization information at least comprises region position and region scale parameters, and additionally outputting region boundary information when the region boundary meets the set condition.
4. The intelligent detection method for defects of the ultrathin electrolytic copper foil oriented to the new energy field, which is characterized in that the space-time alignment is carried out on the suspected defect area set obtained by visual detection, and an electrical verification task queue is generated according to the coil operation process, specifically: Aiming at the suspected defect area set, establishing a space-time mapping relation between an image coordinate and an electrical probe coordinate system based on a coiled material running track, obtaining an expected trigger position and a trigger time window of each suspected defect area at an electrical probe, and taking the expected trigger position and the trigger time window as space-time mapping results; generating an electrical verification task for each suspected defect area based on the space-time mapping result, wherein the electrical verification task at least comprises a trigger position, a trigger time window and mode information related to electrical measurement; Organizing a plurality of electrical verification tasks into an electrical verification task queue according to arrival sequence and/or a preset scheduling strategy in the running process of the coiled material; And scheduling the electrical verification task queue based on the available processing capacity of the electrical probe in the triggering time window and the allowed mode switching constraint in unit time, and outputting the execution sequence and the processing strategy of each task so that the electrical probe triggers local measurement on the suspected defect area in the corresponding triggering time window.
5. The intelligent detection method of the ultrathin electrolytic copper foil defect oriented to the new energy field, which is disclosed in claim 1, is characterized in that the method is used for generating an electrical measurement strategy aiming at an electrical verification task to be triggered, and specifically comprises the following steps: Determining an electricity verification evidence target based on defect type initial judgment, visual confidence and multi-feature description vectors of a suspected defect region, and searching in a preset measurement mode selection strategy library according to the electricity verification evidence target to obtain candidate electricity measurement modes; combining the trigger position and trigger time information determined by the space-time mapping result, and performing executable screening on the candidate measurement modes according to the available resource constraint parameters of the electrical probe; Selecting a corresponding measurement mode from the executable measurement modes based on the discrimination capability evaluation result of the electrical verification evidence target as an electrical measurement mode of the suspected defect area; Determining a scanning path form according to the space form and the equivalent scale of the suspected defect area, and configuring sampling points and point distances based on the relation between the scale of the suspected defect area and the sensitive footprint of the electrical probe so as to meet a preset measurement coverage criterion; And configuring residence time or integration time according to the electrical verification evidence target and the field noise condition so as to enable the measurement uncertainty to be in a discriminable range, and configuring excitation frequency, frequency band or sampling rate by combining a physical mechanism of an electrical measurement mode to form an electrical measurement strategy for driving an electrical probe to perform verification.
6. The intelligent detection method for defects of the ultrathin electrolytic copper foil facing the new energy field as claimed in claim 5, which is characterized in that: The candidate electrical measurement mode further comprises a data-driven selection or layering decision, wherein the data-driven selection or layering decision is obtained by analyzing the associated data of the historical visual characteristics and the electrical verification result, and the layering decision is obtained by performing quick screening on the suspected region and triggering the fine detection quantity mode according to the screening result.
7. The intelligent detection method for defects of the ultrathin electrolytic copper foil facing the new energy field, which is characterized in that the method triggers local electrical measurement by an electrical detection probe according to a generated electrical measurement strategy to obtain local measurement data of a suspected region, specifically comprises the following steps: Triggering local electrical measurement on a designated probe channel and a scanning path according to an electrical measurement strategy in a triggering time or time interval corresponding to the suspected defect area so as to obtain local measurement data aiming at the suspected defect area; constructing a local reference standard at a neighboring position of the same coiled material, and calculating a difference characteristic representing the abnormality degree of the suspected defect area based on the local measurement data and the local reference standard; Carrying out structural convergence and physical consistency verification on the difference features according to the suspected defect region set to generate an electrical confidence coefficient for representing the credibility degree of the structural abnormality; and forming a combined evidence by the electrical confidence and the visual confidence obtained in the visual detection stage, and performing verification and confirmation on the suspected defect area.
8. The intelligent detection method for defects of ultra-thin electrolytic copper foil facing new energy field of claim 7, wherein triggering local electrical measurement on a designated probe channel and scanning path according to electrical measurement strategy further comprises adopting a combined measurement mode, specifically: Firstly, performing rapid screening measurement with low resource occupation to obtain a preliminary electrical response, and when the preliminary electrical response meets a preset upgrading criterion, performing precise detection measurement with higher resolution; And when the difference characteristic and the measurement mode mechanism do not meet the physical consistency check, marking the corresponding suspected defect area as an insufficient evidence state, and triggering one or a combination of sampling supplement, measurement mode upgrading or delayed rechecking.
9. The method for intelligently detecting defects of ultrathin electrolytic copper foil for new energy field as set forth in claim 7, wherein the method for judging suspected defect areas after electrical measurement comprises the following steps: Aiming at a suspected defect area subjected to electrical fine inspection, respectively constructing visual characteristic sub-vectors comprising area geometric forms, texture disturbance characteristics and visual confidence coefficients, and electrical characteristic sub-vectors comprising resistance or conductivity deviation, thickness or morphology deviation and electrical confidence coefficients; fusing the visual feature sub-vector and the electrical feature sub-vector according to a preset structure and performing standardization processing to form a fused feature vector which keeps the distinguishing property of the feature sources; Based on the fusion feature vector execution result, distinguishing and verifying defect types according to the combination relation of the visual confidence coefficient and the electrical confidence coefficient: When the visual confidence coefficient and the electrical confidence coefficient meet the corresponding confirmation threshold values, judging the suspected defect area as a structural defect area; when the visual confidence coefficient meets a confirmation threshold value and the electrical confidence coefficient is lower than the electrical threshold value, judging the suspected defect area as an unstructured abnormal area or marking the suspected defect area as an apparent abnormality to be subjected to degradation treatment; when the visual confidence coefficient is lower than the visual threshold value and the electrical confidence coefficient meets the confirmation threshold value, judging the suspected defect area as a potential concealed defect area or marking the suspected defect area as a concealed abnormality to be rechecked and treated; When both the visual confidence and the electrical confidence are below the corresponding thresholds, the region is determined to be a background region and either discarded or only the low priority records are retained, depending on the scheduling policy.
10. The system for applying the intelligent detection method for the defects of the ultrathin electrolytic copper foil facing the new energy field, which is characterized by comprising a suspected defect area determining module, an electrical verification task queue generating module, an electrical measurement strategy generating module and a suspected result judging module; the module for determining the suspected defect area is used for continuously collecting an original image sequence with fixed line frequency when the ultrathin electrolytic copper foil passes through the field of view of an industrial camera, preprocessing an original image, and visually detecting the preprocessed image to judge to obtain a suspected defect area set; The electricity verification task queue generating module is used for carrying out space-time alignment on the suspected defect area set obtained by visual detection, generating an electricity verification task queue according to the coiled material operation process, and simultaneously executing scheduling on the task queue; the electrical measurement strategy generation module is used for generating an electrical measurement strategy aiming at an electrical verification task to be triggered, wherein the electrical measurement strategy comprises a selected electrical measurement mode, a set electrical scanning path and points; The suspected result judging module is used for triggering local electrical measurement according to the generated electrical measurement strategy through the electrical detection probe, obtaining local measurement data of a suspected region and judging a suspected result of a suspected defect region where the electrical measurement is completed.

Description

Intelligent detection method and system for defects of ultrathin electrolytic copper foil facing new energy field Technical Field The invention relates to the technical field of intelligent detection, in particular to an intelligent detection method and system for defects of an ultrathin electrolytic copper foil in the field of new energy. Background The ultra-thin electrolytic copper foil is an ultra-thin copper material prepared by an electrolytic deposition process, is in a coiled form, has the thickness of about 4-12 mu m in general, has the characteristics of high purity, high conductivity, high ductility, smooth surface and the like, is used as a bearing matrix of a cathode active material in the new energy field, effectively transmits electrons to an external loop, is thinner, can be loaded with more active materials under the same space so as to improve the energy density of a battery, is more likely to evolve into micro-short risks or performance attenuation after a battery system is overlapped due to pinhole, inclusion and local thickness abnormality, is crucial for intelligently detecting defects of the ultra-thin electrolytic copper foil, and aims to discover and quantify related defects of the surface, edge, appearance and thickness of the copper foil on a high-speed production line of continuous production and slitting and rewinding of the copper foil, convert the defects into risk classification and on-line handling capacity facing battery results, detect various defects on the high-speed copper foil production line in real time, accurately classify, locate and label various micron-level defects and reduce the safety accident risks caused by the materials. For example, the invention patent publication No. CN113325001B discloses an automatic distinguishing and detecting device and method for surface appearance flaws of copper foil, the device is arranged between a copper foil processing end and a slitting end, and comprises a CCD linear array camera which is arranged above and below a copper foil production line and is used for synchronously scanning the front and the back of the copper foil in real time, a visual detection image processing system for analyzing and judging the surface image characteristics of the copper foil, a light source generating device and a grading judgment module for grading and classifying the surface flaws according to the sizes of the flaws of the copper foil, the number of the flaws of the copper foil with specified unit length and the degree of flaw density, and the visual detection image processing system is used for identifying whether pinholes, spots, chemical marks and pit pressing phenomena exist on the top surface and the bottom surface of the copper foil. The copper foil defect detection system comprises an acquisition module, a processing module, a matching module and a comparison module, wherein the acquisition module is used for acquiring an initial image corresponding to a copper foil to be detected, the processing module is used for carrying out image preprocessing on the initial image to obtain a standard initial image, the matching module is used for carrying out contour matching on the standard initial image to obtain a detection image of the copper foil to be detected, the comparison module is used for generating a defect label associated with the copper foil to be detected based on the fact that the difference coefficient of the detection image and the standard copper foil image is larger than a threshold value, the detection image of the copper foil to be detected can be extracted, and the detection image and the standard copper foil image are compared, and in the comparison process, a pixel unit of the detection image is taken as a minimum comparison unit. However, in the process of implementing the technical scheme of the embodiment of the application, the application discovers that the above technology has at least the following technical problems: In the prior art, for ultra-thin electrolytic copper foil applied to new energy, optical imaging visual detection is generally adopted to detect the surface of the copper foil on line in the prior art, the method generally takes gray scale, texture, color and local statistical disturbance thereof as main discrimination basis, and can generate higher response to appearance defects such as pinholes, scratches, wrinkles and the like, however, in the high-speed continuous production process of the ultra-thin electrolytic copper foil, the defect risk is not only from apparent appearance abnormality, but also comprises functional defects such as internal thinning, local conductivity abnormality, uneven metallurgical structure, material continuity damage and the like, and the defects often present weak contrast, unstable or insignificant visual characteristics at the imaging level, so that visual detection is difficult to provide evidence directly corresponding to the electrical performance risk, thereby c