Lithium Ion Battery
Lithium-ion batteries are crucial for numerous applications, and research focuses on improving their performance, lifespan, and safety. Current efforts concentrate on developing accurate and robust methods for predicting battery state-of-health (SOH), remaining useful life (RUL), and capacity, employing machine learning models such as transformers, graph neural networks, and various recurrent neural networks, often enhanced by data augmentation and physics-informed approaches. These advancements are vital for optimizing battery management systems, enabling more efficient energy storage and utilization in electric vehicles, grid-scale energy storage, and portable electronics.
Papers
A Comparison of Baseline Models and a Transformer Network for SOC Prediction in Lithium-Ion Batteries
Hadeel Aboueidah, Abdulrahman Altahhan
Fast State-of-Health Estimation Method for Lithium-ion Battery using Sparse Identification of Nonlinear Dynamics
Jayden Dongwoo Lee, Donghoon Seo, Jongho Shin, Hyochoong Bang