The performance and lifespan of lithium-ion batteries are critical factors that determine the overall reliability and cost-effectiveness of the systems they power. To effectively monitor and manage these factors, 'Health Indicators (HIs)' are used, which can predict the State of Health (SOH) and Remaining Useful Life (RUL) of the batteries.
Direct vs Indirect Health Indicators
Direct HIs, such as battery capacity and internal resistance, offer a high degree of accuracy when estimating the SOH and RUL of a battery. However, their real-world application is met with several challenges, including complexities in online monitoring and real-time data acquisition. For example, electrochemical impedance spectroscopy, used to measure internal resistance, is particularly time-consuming and complicated.
Conversely, Indirect HIs, which are constructed from easily-measurable external battery data, offer a practical solution to the challenges posed by direct HIs. They can be obtained online, thereby effectively characterizing battery aging. However, they have their limitations, as the data obtained during the battery discharge stage can be influenced by individual usage habits and environmental factors, limiting their accuracy.
To mitigate these drawbacks, it is recommended ''to derive indirect HIs from data collected during the battery charging stage''. This stage, during which the car is stationary and the battery isn't providing energy externally, is less susceptible to external factors. Consequently, HIs derived in this manner are more stable and accurate for predicting SOH and RUL.
The insightful research by Zhou, Lu, and Zheng presented a comprehensive review of 'Direct and indirect HIs' and introduced the concept of 'Multi-indicator Fusion' for enhancing prediction accuracy in lithium-ion battery aging models.
The Multi-Indicator Fusion
By combining multiple HIs using methods such as grey relational analysis and the entropy weight method, prediction accuracy was significantly improved. Furthermore, various correlation methods ensured the retention of as much original information as possible while reducing redundancy.
The study concluded that indirect HIs, derived from external battery data during the charging stage and combined with multi-indicator fusion techniques, provide a robust and practical approach to predicting battery aging. These methods, the researchers suggest, are more suitable for practical applications than SOH and RUL estimations based on direct health indicators.
To know more, refer to research paper: Zhou W, Lu Q, Zheng Y. Review on the Selection of Health Indicator for Lithium Ion Batteries. Machines. 2022; 10(7):512.
Comments