Advancing battery state of charge estimation in electric vehicles through deep learning: A comprehensive study using real-world driving data

Mohd Herwan, Sulaiman and Zuriani, Mustaffa and Saifudin, Razali and Mohd Razali, Daud (2024) Advancing battery state of charge estimation in electric vehicles through deep learning: A comprehensive study using real-world driving data. Cleaner Energy Systems, 8 (100131). pp. 1-9. ISSN 2772-7831. (Published)

Preview

Pdf
Advancing battery state of charge estimation in electric vehicles.pdf
Available under License Creative Commons Attribution Non-commercial.
Download (3MB) | Preview

DOI/Official URL: https://doi.org/10.1016/j.cles.2024.100131

Abstract

Accurately estimating the State of Charge (SOC) in Electric Vehicles (EVs) is critical for battery management and operational efficiency. This paper presents a Deep Learning (DL) approach to address this challenge, utilizing Feed-Forward Neural Networks (FFNN) to estimate SOC in real-world EV scenarios. The research used data from 70 driving sessions with a BMW i3 EV. Each session recorded key factors like voltage, current, and temperature, providing inputs for the DL model. The recorded SOC values served as outputs. We divided the dataset into training, validation, and testing subsets to develop and evaluate the FFNN model. The results demonstrate that the FFNN model yields minimal errors and significantly improves SOC estimation accuracy. Our comparative analysis with other machine learning techniques shows that FFNN outperforms them, with an approximately 2.87 % lower root mean square error (RMSE) compared to the second-best method, Extreme Learning Machine (ELM). This work has significant implications for electric vehicle battery management, demonstrating that deep learning methods can enhance SOC estimation, thereby improving the efficiency and reliability of EV operations.

Item Type:	Article
Additional Information:	Indexed by Scopus
Uncontrolled Keywords:	Battery; Deep learning; Feed-forward neural networks (FFNN); State of charge
Subjects:	T Technology > TK Electrical engineering. Electronics Nuclear engineering
Faculty/Division:	Faculty of Computing Faculty of Electrical and Electronic Engineering Technology
Depositing User:	Mrs Norsaini Abdul Samat
Date Deposited:	14 Aug 2024 03:45
Last Modified:	14 Aug 2024 03:45
URI:	http://umpir.ump.edu.my/id/eprint/42347
Download Statistic:	View Download Statistics

Actions (login required)

View Item