Battery Separator Black Spot Inspection With Deep Learning Imaging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Visual inspection of rechargeable battery separators for black spots, which cause dV defects due to metal foreign material oxidation, is inefficient and prone to human error, leading to low detection rates and high dispersion among inspectors.
Innovation Solution
A system utilizing cameras and deep learning to automatically detect and analyze black spots on separators, including the removal of foreign materials using adhesion and removal rollers, and X-ray fluorescence analysis for component identification, thereby improving detection accuracy and reducing human error.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If visual inspection with naked eye is used to find black spots, then the method is simple and requires no complex equipment, but the detection precision is low and dispersion among inspectors is high
Solution Approach 1:
The patent replaces the mechanical visual inspection system with an automated optical inspection system using cameras and image processing software. The system captures images of the separator surface and uses algorithmic analysis to detect black spots, eliminating human subjectivity and improving detection precision while maintaining operational simplicity.
Solution Approach 2:
The patent creates a digital copy (image) of the separator surface for analysis. By capturing the separator surface as an image and analyzing it through software, the system allows multiple inspectors to analyze the same data without dispersion, while the image copying enables repeated analysis without additional physical inspection effort.
2Reliability
If multiple cells are dismantled and analyzed by several inspectors, then more black spots may be found, but the time consumption increases and detection efficiency remains low
Solution Approach 1:
The patent enables continuous inspection by processing multiple cells sequentially through the automated system without interruption. The camera system continuously captures images and the software continuously analyzes them, maintaining a steady flow of inspection activity that eliminates the start-stop nature of manual inspection and improves overall throughput.
Solution Approach 2:
The inspection system performs self-analysis through automated image processing algorithms that independently identify black spots without requiring human intervention for each cell. The software automatically processes images, detects defects, and generates results, enabling the system to serve itself and dramatically reducing the time per inspection while maintaining high reliability.
3Measurement precision
If automated equipment is used to detect black spots, then detection precision and consistency improve, but device complexity and initial cost increase
Solution Approach 1:
The patent divides the inspection system into distinct functional modules: image capture (camera), image processing (software), and result analysis (display system). This segmentation allows each component to be optimized independently and enables flexible configuration based on inspection requirements, reducing overall system complexity while maintaining high detection precision.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The system significantly increases the detection ratio of black spots, reduces dispersion among analysts, and enhances inspection speed by using camera imaging and deep learning to select and display the positions of black spots, while automatically analyzing their components with XRF equipment.
Implementation Method 1
the foreign material on the first surface may be removed using the viscosity difference between the first adhesion roller and the first removal roller
Implementation Method 2
X-ray fluorescence analysis for component identification
Data Source
AI summary
A method of finding black spots in a separator according to an embodiment includes taking out a separator from a rechargeable battery cell; removing a foreign material of the separator surface; obtaining a first image for a portion where black spots are estimated in the separator by using a first camera and recording a position of the first image; first selecting the part where black spots are estimated by using the first image, and acquiring a second image for black spots and a foreign material other than black spots in the separator by using a second camera for the recorded position; and secondary selecting black spots by deep learning the first image and the second image with a deep learning software and then displaying the position of the black spots after.