Battery Cell Surface Inspection
Check cell coating quality with deep learning solutions
Graphical programming environment for deep learning-based industrial image analysis
After they are welded, battery cells are wrapped in a durable protective coating. This coating can have flaws, including bubbles and inclusions under the coating, scratches through the coating, and inadequately applied coating. When these cells are packed tightly into a battery module, several factors can lead to an electrical short or overheating: the close proximity of the cells, the charge each cell has to carry, the heat generated by the cells, or inadequate contact with thermal interface material (TIM).
Battery cell coatings can have a variety of minor blemishes that do not compromise function as well as seemingly minor scratches that render them unsafe or unusable. It is important to detect these defects while minimizing rejection of flawed but functional coatings.
The battery cell surface can be inspected by a more sophisticated machine vision systems such as the In-Sight D900 series, which has deep learning inspection capabilities embedded in the vision system.
Cognex Deep Learning is trained with an image set of both good and defective surfaces. Cognex Deep Learning’s defect detection tool learns to identify and pass surfaces within the acceptable range of variation and flags those with unacceptable defects, accounting for natural variations within the image such as light reflection.