Issue |
E3S Web Conf.
Volume 601, 2025
The 3rd International Conference on Energy and Green Computing (ICEGC’2024)
|
|
---|---|---|
Article Number | 00033 | |
Number of page(s) | 11 | |
DOI | https://doi.org/10.1051/e3sconf/202560100033 | |
Published online | 16 January 2025 |
Comparative Analysis of Battery State of Charge Estimation Methods
A2(IS) Research Team, Physics Department, Faculty of Sciences Semlalia, Cadi Ayyad University, Marrakesh, Morocco
* Corresponding author: a.margal.ced@uca.ac.ma
Lithium-ion batteries are widely used in electric vehicles, buses, etc., due to their high-power density, long lifespan, and high energy density. To efficiently manage energy in these vehicles, a Battery Management System (BMS) is crucial. A critical parameter for the BMS is the State of Charge (SoC), which indicates the available charge in the battery and ensures its operational range. This paper presents three methods for estimating SoC: the extended Kalman filter (EKF), the adaptive Luenberger observer (ALO), and a neural network model employing nonlinear auto-regressive with eXogenous inputs (NARX). These methods are evaluated under the LA92 driving cycle using metrics like Root Mean Square Error (RMSE) to assess their performance. Results show that the NARX model achieves the highest accuracy with an RMSE of 0.33%, followed by the EKF with 5.34% and finally the ALO with 5.94%. These findings indicate that all three methods are acceptable, and the proposed NARX model shows superior performance. With the NARX model exhibiting superior performance in SoC estimation for electric vehicle applications.
© The Authors, published by EDP Sciences, 2025
This is an Open Access article distributed under the terms of the Creative Commons Attribution License 4.0, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.