Lithium-ion Batteries State of Charge Comparison between Extended Kalman Filter and Machine Learning

The surge in electric vehicle popularity has significantly driven the growth of the lithium-ion battery sector, emphasizing the need for technological innovations that ensure safe and efficient battery management to maximize energy utilization. This study introduces novel methodologies for accurately determining the State of Charge (SoC) of lithium-ion batteries, with a focus on a Lithium Nickel-Cobalt-Aluminum Oxide (NCA) battery in a commercial vehicle. It explores two SoC estimation strategies: Extended Kalman Filters (EKF) and Machine Learning, with a special focus on the latter due to its innovative approach and potential efficiency in Battery Management Systems (BMS). This research aims to improve the precision of SoC estimations, pro-viding insights into the effectiveness of these methodologies, their unique contributions to the field, and their wider implications for enhancing battery management in electric vehicles. Through a comparative analysis and an emphasis on the capabilities of machine learning, the study seeks to deepen the understanding of battery performance and management, paving the way for further advancements in electric vehicle efficiency and safety.