Work

Smart Artificial Neural Network Model With an Optimization Strategy for State of Charge Estimation of Lithium-Ion Battery Packs in Electric Vehicles

Year: 2026
Type: conference-paper
Abstract: Accurate prediction of the state of charge (SoC) is critical to the proper functioning of lithium-ion cells in electric vehicles. As a key indicator of remaining energy relative to battery capacity, SoC is essential to guarantee safe operation and prevent any deterioration. This study proposes a systematic artificial neural network (ANN) framework with an optimized configuration strategy, resulting in enhanced accuracy and robustness in SoC estimation for li-ion cells. The model is trained and validated using data generated from an electric vehicle battery pack. The considered system consists of a battery pack designed for EV applications, composed of multiple li-ion cells connected in series. The performance of the proposed approach is evaluated under CC-CV charging and constant-current (CC) discharging conditions to enhance model robustness and predictive accuracy. During the experiments, key parameters including voltage, current, and temperature are systematically monitored. Performance is assessed using standard evaluation metrics. The results demonstrate that the proposed model achieves high accuracy, with a RMSE of 0.978 %, a MAE of 0.615 %, and a$\mathbf{R}^{\mathbf{2}}$of 0.9994.
Language: English
Open Access status: hybrid
OpenAlex ID: W7166520122
FWCI:
Cites: 27
Cited by:
Related to: 10