IEEE Transaction on Industrial Informatics (2022)
Zihan Zhang(Georgia Institue of Tech), Yeonjeong Jeong, Jongseong Jang, Chi-ghun Lee(University of Toronto)
Abstract
Recently, there has been a significant growth in the development of rechargeable battery-powered devices, leading to an urgent need for reliable and safe batteries. The end-of-life (EoL) is an important measure for battery and defined as the number of charge and discharge cycles before battery health becomes unacceptably bad. The EoL can be estimated by adaptive stochastic processes or advanced machine learning techniques. However, the existing approaches either assume over-simplified degradation pattern or act as a black box offering no interpretation. To address these limitations, we develop a pattern-driven degradation process by integrating a recursive Gaussian distribution with its mean learnt from an GRU-driven degradation pattern. Due to the non-Markovian state transitions, a joint-learning sampling-based expectation maximization algorithm was developed to estimate model parameters. Finally, numerical studies showed that the proposed method achieves over 3%/40% higher accuracy in EoL prediction than the GRU/adaptive Wiener process, respectively.