Battery Cell Tab Defect Detection for Folding-to-Film Risk
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional tab defect detection methods in battery cell production fail to accurately distinguish between minor defects like folding or wrinkling and those that pose a risk of folding into the film region, leading to reduced yield and increased manual reevaluation.
Innovation Solution
A method and device that utilize two-sided imaging of tab stacks, employing AI-based algorithms to detect tab line abnormalities and determine the position of defects, thereby assessing the risk of folding into the film region by combining detection results from both sides, ensuring only severe defects are flagged as unqualified.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional single-sided tab defect detection is used, then the detection process is simple, but the accuracy of distinguishing minor defects from serious defects is low
Solution Approach 1:
The detection system is segmented into two separate detection paths: one for detecting first-type defects (tab line abnormalities) and another for detecting second-type defects (folding-to-film region defects). This segmentation allows each detection path to be optimized for its specific defect type, improving overall detection accuracy while maintaining manageable system complexity through modular design.
Solution Approach 2:
The invention transitions from single-sided detection to two-sided detection by adding detection of the opposite side of the tab stack. This dimensional change provides complementary information that enables more accurate distinction between minor defects (visible on one side only) and serious defects (visible on both sides), significantly improving measurement precision.
2Reliability
If all detected defects are flagged as unqualified, then defect detection sensitivity is high, but cell yield decreases due to false positives
Solution Approach 1:
The system uses feedback from two-sided detection results to dynamically adjust defect classification. By comparing defects detected on both sides, the system provides feedback that helps distinguish between minor defects (present on only one side) and serious folding-to-film defects (present on both sides), improving reliability while reducing false positives that would otherwise reduce yield.
Solution Approach 2:
Different quality criteria are applied to different defect types based on their detection characteristics. First-type defects (tab line abnormalities) and second-type defects (folding-to-film defects) are evaluated with different thresholds and classification rules, allowing the system to maintain high reliability for serious defects while being more lenient with minor defects, thereby preserving cell yield.
3Measurement precision
If manual reevaluation is increased to verify defects, then detection accuracy improves, but workload and time consumption increase
Solution Approach 1:
The detection system performs self-service by automatically classifying defects based on two-sided detection comparison. The system independently determines whether a defect is minor or serious without requiring manual intervention, achieving high classification accuracy through automated image processing and comparison algorithms, thereby eliminating time-consuming manual reevaluation.
Solution Approach 2:
The system performs preliminary classification of defects automatically before any potential manual review. By pre-processing and classifying defects based on two-sided detection data, the system prepares qualified results in advance, minimizing the need for subsequent manual verification and reducing overall processing time while maintaining high accuracy.
Data Source
AI summary
Disclosed is a method and device for detecting defects in battery cell tabs. The method includes acquiring a first image of one side surface and a second image of the opposite side surface of a tab stack, where both surfaces are parallel to the stacking direction and extend along the height of the tabs. Based on the first and second images, detection results are obtained indicating whether a tab has a first-type defect and identifying the defect's position relative to the tab height. The first-type defect includes tab line abnormalities. The method further determines, from the detection results, whether a folding-to-film region defect is present in the tab stack. This approach improves accuracy in detecting tab folding defects, enabling better assessment of battery cell quality.