Air-Coupled Ultrasound Spectroscopy for Electrolyte Fill Level Detection in Lithium-Ion Pouch Cell Batteries

Current testing of Li-ion pouch cell batteries in manufacturing settings is performed through batch testing and often by destructive means. This process is costly and is insensitive to some defects. Detecting defects earlier in the production process than the end-of-line is imperative for financial and material efficiency and safety objectives to be met. Internal defects, including an improper fill-level of electrolyte, commonly result in explosive thermal runaway and fire, which pose a significant danger to human life. Therefore, the development of methods to rapidly and robustly detect internal defects within cells assists in the initial refinement of production and allows for the identification of defects to prevent these batteries from entering the consumer market. Innovative analysis of signals from air-coupled ultrasound scans for batteries with different electrolyte fill-levels, uses a combination of signal metrics and wavelet transform analysis. Results of Support Vector Machine classification models are used to demonstrate the potential for predicting anomalous fill-levels from air-coupled ultrasound signals, conducive to rapid in-line inspection in gigafactories.