As the new energy vehicle sector advances at an accelerated pace, lithium-ion batteries have grown far more prevalent across electric vehicles, energy storage systems, and portable electronic devices. The State of Health (SOH) of lithium batteries bears direct implications for their safety, performance stability, and operational lifespan. For this reason, Prognostics and Health Management (PHM) technology tailored to lithium batteries has steadily become a key area of focus in both academic inquiries and industrial applications. This paper systematically reviews the research progress of lithium battery PHM technology in recent years, mainly covering key methods such as battery thermal state characterization indicators, Physics-Informed Neural Network (PINN), and Integrated Sparse Gaussian Process Regression (SGPR). This paper not only summarized the core principles and applications of each technology, but also analyzed its shortcomings and proposes several improvement directions. This paper provided a reference for future research on SOH prediction and health management of lithium-ion batteries.
Research Article
Open Access