基于递推最小二乘法的锂离子电池等值参数在线识别  

On⁃line Identification Parameters of Lithium⁃ion Battery Equivalent Model Based on Recursive Least Square Method

作　　者：李泽宇 王晓磊 张世彪 石淼岩 LI Zeyu;WANG Xiaolei;ZHANG Shibiao;SHI Miaoyan(School of Electrical Engineering,Shandong University,Jinan 250061,China;Huaneng Shandong Power Generation Co.,Ltd.,Jinan 250014,China;State Grid Zibo Power Supply Company,Zibo 255000,China)

机构地区：[1]山东大学电气工程学院,山东济南250061 [2]华能山东发电有限公司,山东济南250014 [3]国网山东省电力公司淄博供电公司,山东淄博255000

出　　处：《山东电力技术》2021年第2期1-6,16,共7页Shandong Electric Power

基　　金：国家自然科学基金项目(51707108);山东省国家自然科学基金资助项目(2017QEE015)。

摘　　要：电池状态的准确在线估计是锂离子电池储能系统安全、可靠和高效运行的重要支撑,精确电池模型是实现上述估计的前提。然而,锂电池复杂的电化学反应过程使电池模型参数可能随荷电状态(State of Charge,SOC)、温度及运行工况改变而变化,传统的离线识别方法难以保证模型精度和适应性。提出一种基于递推最小二乘法(Recursive Least Squares,RLS)的在线参数识别方法,该方法能够自动适应电池SOC、温度及运行条件变化,拟合计算出最佳模型参数,进而保证模型精度。研究不同电池类型下等值模型结构对建模精度的影响,并通过混合脉冲功率性能测试试验,验证了方法有效性。算例分析结果表明:基于RLS的在线参数识别方法能够实时修正电池模型参数,提高电池模型建模精度;对于磷酸铁锂电池,采用二阶RC模型可比一阶RC模型获得更高的建模精度。Accurate on-line operating state estimation is an important support for the safe,reliable and efficient operation of lithiumion battery energy storage system,and an accurate battery model is the premise to realize the estimation.However,due to the complex electrochemicalreaction process of lithium battery,the battery model parameters may change with SOC,temperature and operating conditions.The traditional off-line identification methods are difficult to ensure the accuracy and adaptability of the model.An on-line parameter identification method based on RLS was proposed.The method can automatically adapt to the changes of SOC,temperature and operating conditions,fit and calculate the best model parameters,and then ensure the accuracy of the model.The influence of equivalent model structure on modeling accuracy under different battery types was studied,and the effectiveness of the method was verified by mixed pulse power performance test.The results show that the online parameter identification method based on RLS can correct the battery model parameters in real-time and improve the modeling accuracy of battery model.In addition,for phosphate iron lithium battery,the second-order RC model can obtain higher modeling accuracy than the first-order RC model.

关 键 词：参数在线识别 锂离子电池等效电路 递推最小二乘法(RLS) 混合脉冲功率性能试验 

分 类 号：TM912[电气工程—电力电子与电力传动]