Optimizing Lithium-Ion Battery Performance and Safety for E-Bikes: A Review of Machine Learning-Driven Battery Management Systems

The growing demand for sustainable transportation has positioned electric bikes (e-bikes) as a key solution, with lithium-ion batteries (LIBs) critical for their performance and reliability. This review provides a comprehensive examination of recent advancements in optimizing LIBs for e-bikes, focusing on integrating machine learning (ML) into Battery Management Systems (BMS) and developing fast-charging solutions. The review explores state-of-the-art machine learning models used for State of Charge (SOC) and State of Health (SOH) estimation, significantly improving prediction accuracy, adaptability, and battery safety under real-world conditions. Fast-charging technologies, essential for enhancing the user experience of e-bikes, are also evaluated, focusing on balancing rapid charging and minimizing degradation. Despite these advancements, challenges remain in real-time system integration, computational efficiency, and thermal management. The review highlights future research opportunities, including developing lightweight, adaptive AI models and novel materials for improving energy density and safety. This work aims to advance the design and application of LIBs in e-bikes, contributing to the broader adoption of sustainable, electric transportation.