期刊
JOURNAL OF POWER SOURCES
卷 243, 期 -, 页码 805-816出版社
ELSEVIER
DOI: 10.1016/j.jpowsour.2013.06.076
关键词
Lithium-ion battery; Data driven; Dynamic universal battery model; Adaptive extended Kalman filter; State of charge
资金
- US DOE Grant [DE-EE0002720, DE-EE0005565]
- Graduate School of Beijing Institute of Technology in part
- Higher education innovation intelligence plan (111plan) of China
This paper presents a novel data-driven based approach for the estimation of the state of charge (SoC) of multiple types of lithium ion battery (LiB) cells with adaptive extended Kalman filter (AEKF). A modified second-order RC network based battery model is employed for the state estimation. Based on the battery model and experimental data, the SoC variation per mV voltage for different types of battery chemistry is analyzed and the parameters are identified. The AEKF algorithm is then employed to achieve accurate data-driven based SoC estimation, and the multi-parameter, closed loop feedback system is used to achieve robustness. The accuracy and convergence of the proposed approach is analyzed for different types of LiB cells, including convergence behavior of the model with a large initial SoC error. The results show that the proposed approach has good accuracy for different types of LiB cells, especially for C/LFP LiB cell that has a flat open circuit voltage (OCV) curve. The experimental results show good agreement with the estimation results with maximum error being less than 3%. (C) 2013 Elsevier B.V. All rights reserved.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据