Real-time state-of-charge estimation for rechargeable batteries based on in-situ ultrasound-based battery health monitoring and extended Kalman filtering model

Ultrasonic testing has emerged as a crucial non-invasive method for monitoring battery health, particularly for accurate State-of-Charge (SoC) estimation in Battery Management Systems (BMS). Unlike invasive methods relying on real-time collection of battery current and voltage, ultrasonic inspection offers timely feedback without interfering with battery properties. However, challenges remain in accurately estimating SoC during rechargeable battery discharging due to ultrasonic echo interference. This study presents an ultrasound-based in- situ rechargeable battery health monitoring system, incorporating advanced signal processing techniques. The proposed Ultrasonic Signal Empirical Mode Decomposition-Extended Kalman Filtering (USED-EKF) algorithm, based on Biot's theory, achieves real-time SoC estimation with exceptional accuracy (maximum error 0.63 %). Compared to conventional EKF, USED-EKF outperforms with significantly lower errors under constant current conditions. Additionally, our model enables the detection of overcharged batteries using ultrasound echo for the first time. This research demonstrates the potential of ultrasonic testing in cost-effective battery maintenance and explosion prevention, contributing to advancements in battery monitoring and safety measures. This research showcases the potential of ultrasonic testing as a cost-effective tool for battery maintenance and the prevention of battery explosions. The achieved results position our study as a pivotal driver in expediting these critical processes, highlighting the significance of our proposed model in advancing battery monitoring and safety measures.