AI-based estimation of lithium-ion battery management system: a review of AI integration in electric vehicle / Mohammad Lukman Mohd Yasin, Mahanijah Md Kamal and Kanendra Naidu Vijyakumar

• Main Authors: Mohd Yasin, Mohammad Lukman, Md Kamal, Mahanijah, Vijyakumar, Kanendra Naidu
• Format: Article
• Language: English
• Published: UiTM Press 2024
• Subjects: Back propagation (Artificial intelligence); Dielectric devices
• Online Access: https://ir.uitm.edu.my/id/eprint/105779/1/105779.pdf; https://ir.uitm.edu.my/id/eprint/105779/

Lithium-ion (Li-ion) batteries have gained considerable attention in the Electric Vehicle (EV) industry due to their high energy density, better lifespan, and higher nominal voltage. However, accurately estimating the State of Charge (SOC) and State of Health (SOH) for Li-ion batteries remains challenging due to its aging and nonlinear behaviour. This paper explores Battery Management System (BMS) models potential incorporating Artificial Intelligence (AI) estimation techniques, particularly Deep Learning (DL), to improve SOC and SOH model estimations. This research paper summarized and analyzed current BMS approaches by identify the potential gaps in existing research focus and propose another technique for further exploration in the EV Li-ion battery. Currently, there is a research gap in the existing studies, especially in the application of DL for SOC and SOH estimation. and underscores the need for more comprehensive exploration and refinement of DL methods. Future research should address these gaps to advance the integration of DL into BMS to ensure robust and reliable SOC and SOH estimations. Because of its features and capacity to improve SOC and SOH estimating health models accurately, deep learning has a lot of potential for studying SOC & SOH in BMS. As a result, there is opportunity to investigate the DL technique further in order to thoroughly and clearly examine the correctness of SOC & SOH model estimations in BMS.