| 黄德扬,陈自强,郑昌文.时变温度环境下锂离子电池自适应SOC估计方法[J].装备环境工程,2018,15(12):28-34. HUANG De-yang,CHEN Zi-qiang,ZHENG Chang-wen.SOC Adaptive Estimation Method for Li-Ion Battery Applied in Temperature-varying Condition[J].Equipment Environmental Engineering,2018,15(12):28-34. |
| 时变温度环境下锂离子电池自适应SOC估计方法 |
| SOC Adaptive Estimation Method for Li-Ion Battery Applied in Temperature-varying Condition |
| 投稿时间：2018-08-03 修订日期：2018-12-25 |
| DOI：10.7643/ issn.1672-9242.2018.12.005 |
| 中文关键词: 镍钴锰酸锂电池 荷电状态 变温环境 电池特性 轻度混合动力 极地科考船 |
| 英文关键词:NCM lithium-ion battery state of charge varying ambient temperature battery characteristics mild hybrid power polar scientific expedition ship |
| 基金项目:国家自然科学基金项目（51677119） |
|
|
|
|
| 摘要点击次数: |
| 全文下载次数: |
| 中文摘要: |
| 目的 针对低温环境下锂离子电池特性显著变化问题，为大规模锂离子电池组在极地科考船混合动力系统上的应用提供理论依据，方法 对10 Ah高功率三元镍钴锰酸锂电池低温特性展开实验研究，结合实验数据，利用基于遗忘因子的递推最小二乘算法（FFRLS）分别与两种改进的卡尔曼滤波算法（AEKF、UKF）组成的串联观测器在线估计电池荷电状态（SOC）。结果 在25~?30 ℃时变温度环境的改进DST工况下，FFRLS-AEKF算法的SOC估计精度略高于FFRLS-UKF算法，其最大估计误差为3.04%，均方根误差为0.69%。结论 相比EKF与RLS-EKF算法，更好的模型参数与噪声信息的自适应性使FFRLS-AEKF算方法有更高的SOC估计精度与收敛性。 |
| 英文摘要: |
| Objective Considering the significant change of battery characteristics at low temperature, to provide a theoretical basis for the application of large scale lithium ion battery packs in the hybrid power system of polar scientific expedition ships. Methods The low temperature characteristics of 10 Ah high power NCM lithium-ion battery were experimentally investigated. Combined with the experimental data, the state of charge (SOC) was estimated on line with a series observer, which was composed of recursive least squares algorithm with forgetting factor (FFRLS) and two improved kalman filtering algorithms (AEKF, UKF) respectively. Results The SOC estimated accuracy of FFRLS-AEKF algorithm was slightly higher than that of FFRLS-UKF algorithm under the improved DST condition of time-varying temperature environment within the temperature range of 25 ~?30 ℃, with the maximum estimated error of 3.04% and the root mean square error of 0.69%. Conclusion Compared with EKF and RLS-EKF algorithms, the better adaptability of model parameters and noise information makes FFRLS-AEKF algorithm have higher SOC estimated accuracy and convergence. |
| 查看全文 查看/发表评论 下载PDF阅读器 |
| 关闭 |
