Battery Impedance Inspection for Non-Destructive Electrode Crack Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for identifying cracks in battery electrodes require destructive analysis, which leads to battery disposal and lengthy sample preparation, making them unsuitable for immediate feedback during battery use.
Innovation Solution
A non-destructive battery inspection method using electrochemical impedance spectroscopy to acquire impedance spectral data, followed by function fitting and differentiation to determine the battery state based on a second function, allowing crack detection within seven minutes without disassembly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If battery inspection is performed manually, then flexibility in handling various battery types is maintained, but inspection time and labor costs increase
Solution Approach 1:
The inspection system employs a universal robotic end effector that can handle multiple battery types (prismatic, pouch, cylindrical) through a single platform. The system achieves multi-functionality by equipping the robotic arm with interchangeable end effectors and using vision systems that can recognize and adapt to different battery geometries, eliminating the need for separate inspection lines for each battery type while maintaining high automation levels
Solution Approach 2:
The inspection system is divided into modular functional units including vision inspection modules, robotic manipulation modules, and data processing modules. This segmentation allows the system to handle different battery types by activating specific modules while maintaining overall system automation, thus reducing the complexity burden of universal handling capabilities
2Measurement precision
If traditional vision systems are used for battery inspection, then setup is simple, but inspection precision and defect detection capability are insufficient
Solution Approach 1:
The system merges multiple vision technologies (2D cameras, 3D scanners, hyperspectral imaging) into a unified inspection platform. This combination enables simultaneous capture of surface geometry, color defects, and subsurface anomalies, achieving comprehensive defect detection while the integrated system architecture manages complexity through centralized processing and coordinated operation of all vision components
Solution Approach 2:
The system introduces AI-based image processing algorithms as an intermediary between the multi-modal vision sensors and defect detection. This intermediary layer processes and fuses data from different vision modalities, enhancing detection precision while abstracting the complexity of multi-sensor coordination from the control system
3Reliability
If comprehensive battery inspection is performed, then product quality improves, but inspection time and production cycle lengthen
Solution Approach 1:
The inspection system operates continuously as the robotic arm moves batteries through the inspection zone without stopping. Multiple vision sensors capture images simultaneously during the brief moment the battery is positioned, enabling comprehensive inspection (surface, edges, labels, geometry) to be completed in a single continuous operation rather than through multiple sequential inspection steps
Solution Approach 2:
The system performs preliminary positioning and orientation of batteries using robotic manipulation before inspection begins. This preliminary action ensures optimal viewing angles for all inspection surfaces are achieved automatically, allowing comprehensive inspection to be completed faster without manual intervention during the actual inspection phase
Data Source
Figure 1~2
Figure 3~4
AI summary
The present invention relates to a battery inspection method comprising: a data acquisition step of acquiring impedance spectrum data for a finished secondary battery in order to identify cracks in the electrodes of the secondary battery through non-destructive inspection; a first function acquisition step of acquiring a first function, having the log scale of the measurement frequency as an independent variable and the absolute value of impedance as a dependent variable, from the impedance spectrum data through function fitting; a second function acquisition step of acquiring a second function by differentiating the first function with respect to the log scale of the measurement frequency; and a battery state determination step of determining the state of the secondary battery on the basis of the second function.