Non-Contact Laser Ultrasonic Tech for Real-Time Battery Monitoring

Researchers have developed a non-contact laser excited ultrasonic sensing (LEUS) system combined with a Transformer deep learning model to monitor lithium-ion battery State of Charge (SoC) and State of Health (SoH) in real-time. Published in Science Advances, this method eliminates the need for coupling agents or liquid immersion, which are typically required in traditional ultrasonic testing. This breakthrough addresses the limitations of traditional electrochemical and contact-based ultrasonic monitoring, which struggle under high-rate charging conditions and temperature fluctuations. It offers a non-destructive, highly accurate, and scalable solution for real-time battery diagnostics across various battery chemistries and designs. The system utilizes a dual-laser design that increases ultrasonic signal amplitude by 10 times and achieves a signal-to-noise ratio of 30dB. Validated on over 100,000 samples across LFP and NCM batteries, the Transformer model achieves a mean prediction error of under 5.7% for SoC and under 2.1% for SoH.