Inspection method for lithium secondary battery
An inspection method for a lithium secondary battery can improve the reliability, accuracy, and reproducibility of inspection results by advancing a learning of an artificial neural network according to one or more divided regions of one or more surfaces of the lithium secondary battery and one or more types of defects occurring in each region.
1 . An inspection method for a lithium secondary battery comprising:
a first step of determining a plurality of surfaces of the lithium secondary battery, for performing an appearance inspection, dividing each of the plurality of surfaces into one or more regions for one or more types of defects to be detected and accumulating first image data for each region of the one or more regions;
a second step of constructing a second image data including existing image data in which normal or defective is confirmed according to the one or more types of defects occurring in each region;
a third step of comparing the first and second image data with each other in an artificial neural network to first judge whether the lithium secondary battery is normal or defective as a first judgment result;
a fourth step of judging defect detection or erroneous detection from the first judgement result as a second judgment result, and feeding back the second judgement result to cause the artificial neural network to learn;
a fifth step of determining a third image data in which standards for normal or defect are set and classified feeding back of the second judgment result; and
a sixth step of comparing the first image data for each region of the lithium secondary battery to be inspected with the third image data to determine whether the lithium secondary battery to be inspected is normal or defective.
2 . The inspection method of claim 1 , wherein:
the plurality of surfaces comprises an upper surface, one side surface, and a lower surface of a cylindrical, prismatic, or pouch-type lithium secondary battery.
3 . The inspection method of claim 2 , wherein;
the upper surface of the lithium secondary battery is divided into a first region where a dent defect occurs, a second region where an electrolyte leakage defect and a washer defect occur, and a third region where an electrolyte input defect occurs, and
the first to third image data of the upper surface comprises image data selectively divided according to the first to third regions.
4 . The inspection method of claim 3 , wherein:
the second to fourth steps are performed in parallel or sequentially for each of the upper surface, the lower surface and the one side surface of the lithium secondary battery.
5 . The inspection method of claim 4 , wherein:
the second to fourth steps are performed by the one or more types of defects for each divided region, and
the artificial neural network determines the third image data in which standards for normal or detective with respect to the one or more divided regions are set and classified, based on the learning results for the one or more types of defects.
6 . The inspection method of claim 2 , wherein:
the lower surface of the lithium secondary battery is a fourth region in which a dent defect occurs, a scratch defect occurs, a sink defect occurs, and a surface contamination occurs, and
the first to third image data of the lower surface comprises image data selectively divided according to the fourth region.
7 . The inspection method of claim 6 , wherein:
the second to fourth steps are performed in parallel or sequentially for each of the upper surface, the lower surface and the one side surface of the lithium secondary battery.
8 . The inspection method of claim 7 , wherein:
the second to fourth steps are performed by the one or more types of defects for each divided region, and
the artificial neural network determines the third image data in which standards for normal or detective with respect to the one or more divided regions are set and classified, based on the learning results for the one or more types of defects.
9 . The inspection method of claim 2 , wherein:
the one side surface of the lithium secondary battery is divided into a fifth region in which a dent defect occurs and a sixth region in which a double washer defect occurs, and
the first to third image data of the one side surface comprises image data selectively divided according to the fifth and sixth regions.
10 . The inspection method of claim 9 , wherein:
the second to fourth steps are performed in parallel or sequentially for each of the upper surface, the lower surface and the one side surface of the lithium secondary battery.
11 . The inspection method of claim 10 , wherein:
the second to fourth steps are performed by the one or more types of defects for each divided region, and
the artificial neural network determines the third image data in which standards for normal or detective with respect to the one or more divided regions are set and classified, based on the learning results for the one or more types of defects.
12 . The inspection method of claim 2 , wherein:
the second to fourth steps are performed in parallel or sequentially for each of the upper surface, the lower surface and the one side surface of the lithium secondary battery.
13 . The inspection method of claim 12 , wherein:
the second to fourth steps are performed by the one or more types of defects for each divided region, and
the artificial neural network determines the third image data in which standards for normal or detective with respect to the one or more divided regions are set and classified, based on the learning results for the one or more types of defects.