高倍率脉冲放电锂电池的电气参数辨识  

Electrical parameters identification of high-rate pulse-discharged lithium battery

作　　者：柳应全 王岩冰 郭赟 LIU Yingquan;WANG Yanbing;GUO Yun(National Key Laboratory of Electromagnetic Energy,Naval Univ.of Engineering,Wuhan 430033,China;Beijing Blue Ocean Advanced Technology Co.,Ltd.,Beijing 100176 China;School of Electrical and Electronic Engineering,Huazhong Univ.of Science and Technology,Wuhan 430074,China)

机构地区：[1]海军工程大学电磁能技术全国重点实验室,武汉430033 [2]中航蓝海(北京)国际科技有限公司,北京100176 [3]华中科技大学电气与电子工程学院,武汉430074

出　　处：《海军工程大学学报》2024年第5期41-47,共7页Journal of Naval University of Engineering

基　　金：国家自然科学基金资助项目(52307076,92266202,92266106)。

摘　　要：为了实施准确的电池建模和状态估计,需要对电气参数的辨识方法和精度开展对比研究。因此,在综合对比了现阶段常用的3种参数辨识方法的基础上,提出了一种带参数约束的递归最小二乘法,可实现对输出参数的边界约束,从而避免电气参数出现负值的情况,并以预测电压均方差最小为目标输出最优解。利用4种不同算法对实际高倍率脉冲放电样本数据进行了参数辨识和对比分析,结果表明:所提算法辨识的电气参数比其他智能优化算法精度更高,且运算中的时间消耗大大减少。In order to implement accurate battery modeling and its state estimation,it is necessary to carry out comparative research on the identification method and accuracy of the electrical parameters.Therefore,a constrained recursive least square(CRLS)algorithm was proposed based on the comprehensive comparison of three parameter identification methods commonly used at present.The proposed method can realize the boundary constraint of output parameters so as to avoid the negative va-lue of the electrical parameters,and take the minimum predicted voltage mean square error as the target to obtain the optimal solution.Four different methods were applied to the actual high-rate pulse-discharged sample data,and the parameter identification and comparative analysis,the results show that the electrical parameters identified by CRLS algorithm are more accurate than those by intelligent optimization algorithm,and the time cost is greatly reduced.

关 键 词：高倍率 脉冲放电 参数辨识 电气约束 

分 类 号：TM912[电气工程—电力电子与电力传动]