Comet Yxlon is launching its new automated battery analysis software at Control. This relies on deep learning, which significantly increases accuracy and helps to reduce the reject rate in battery production.

Increasing e-mobility has set the battery market into gigantic growth in recent years, and there is no end in sight. In the same proportion, there is a growing demand for testing technologies that ensure the performance and safety of the batteries, but which should also ensure that the current high level of waste is brought to its knees and the productivity of the manufacturers increased. X-ray technology and especially computed tomography, which can be used to analyze battery cells, modules and entire packages in detail, are best suited for this.

In order for the complex battery cells to be the power packs that are expected, no errors must occur during production. There are many sources of error: breaks, air pockets, foreign bodies, deformations, faulty welds or poor alignment of the inner layers. It becomes dangerous when the overhangs are not sufficient and contact between the cathode and anode causes a short circuit and then a fatal fire. The safety aspect is therefore clearly in the foreground when testing batteries, and so checking and measuring the overhangs has the highest priority. Objective, reliable and repeatable test results can only be obtained with software evaluation.

Software knows no signs of fatigue or sensitivities. The 3D image analysis software Dragonfly relies on deep learning models. The automatic evaluation can be adapted to the specific requirements of the individual manufacturer. The Battery Insights software package, of which Dragonfly is a part, takes care of the fully automatic checking of overhangs and can also process entire batches with the appropriate multi-part holders. The report, which is also generated automatically, provides information about the condition of each individual cell and can be used to identify trends for process optimization in order to significantly reduce rejects.

The Battery Insights software package was presented live at Control with the Comet Yxlon CT system FF35 CT.