TY - GEN
T1 - Battery Life Prediction Using Physics-Based Modeling and Coati Optimization
AU - Safavi, Vahid
AU - Mohammadi Vaniar, Arash
AU - Bazmohammadi, Najmeh
AU - Vasquez, Juan C.
AU - Guerrero, Josep M.
PY - 2024/10/18
Y1 - 2024/10/18
N2 - Accurate remaining useful life (RUL) prediction is essential for ensuring the reliability and efficiency of Lithium-ion (Li-ion) batteries. This paper presents an approach using the Coati Optimization Algorithm (COA) to optimize the physics-based model for RUL prediction of Li-ion batteries. This method combines COA to optimize the physics-based degradation model to improve battery aging predictions, considering factors like cycle time, rest time, temperature, state of charge (SOC), and load conditions. The model can more accurately simulate real-world battery usage patterns and degradation mechanisms by incorporating these variables. Simulation results show that COA enhances the accuracy of the model's calendar and cycle aging prediction, and reduces RMSE and MAE values for RUL prediction. Furthermore, the robustness of the proposed method is demonstrated through extensive testing under various operational scenarios, highlighting its potential for application in battery management systems to extend battery life and improve performance.
AB - Accurate remaining useful life (RUL) prediction is essential for ensuring the reliability and efficiency of Lithium-ion (Li-ion) batteries. This paper presents an approach using the Coati Optimization Algorithm (COA) to optimize the physics-based model for RUL prediction of Li-ion batteries. This method combines COA to optimize the physics-based degradation model to improve battery aging predictions, considering factors like cycle time, rest time, temperature, state of charge (SOC), and load conditions. The model can more accurately simulate real-world battery usage patterns and degradation mechanisms by incorporating these variables. Simulation results show that COA enhances the accuracy of the model's calendar and cycle aging prediction, and reduces RMSE and MAE values for RUL prediction. Furthermore, the robustness of the proposed method is demonstrated through extensive testing under various operational scenarios, highlighting its potential for application in battery management systems to extend battery life and improve performance.
KW - COA optimization
KW - Battery RUL prediction
KW - Physics based model
UR - http://www.scopus.com/inward/record.url?scp=85208239805&partnerID=8YFLogxK
U2 - 10.1007/978-3-031-74741-0_20
DO - 10.1007/978-3-031-74741-0_20
M3 - Article in proceeding
SN - 978-3-031-74740-3
VL - 15272
T3 - Lecture Notes in Computer Science
SP - 303
EP - 313
BT - Energy Informatics - 4th Energy Informatics Academy Conference, EI.A 2024, Proceedings
A2 - Jørgensen, Bo Nørregaard
A2 - Ma, Zheng Grace
A2 - Wijaya, Fransisco Danang
A2 - Irnawan, Roni
A2 - Sarjiya, Sarjiya
PB - Springer Nature Switzerland AG
CY - Kuta, Bali, Indonesia
ER -