基于神经网络的老化锂电池SOC估算方法的研究  

Research on SOC Estimation Method for Aging Lithium Battery Based on Neural Network

作　　者：张立佳[1] 徐国宁[1] 赵向阳[2] 杜晓伟 周翔 ZHANG Lijia;XU Guoning;ZHAO Xiangyang;DU Xiaowei;ZHOU Xiang(Academy of Opto-Electronics,Chinese Academy of Sciences,Beijing 100094,China;School of Automation Science and Electrical Engineering,Beihang University,Beijing 100191,China;Institute of Remote Sensing and Digital Earth,Chinese Academy of Sciences,Beijing 100094,China)

机构地区：[1]中国科学院光电研究院,北京100094 [2]北京航空航天大学自动化科学与电气工程学院,北京100191 [3]中国科学院遥感与数字地球研究所,北京100094

出　　处：《电源学报》2020年第1期54-60,共7页Journal of Power Supply

摘　　要：针对锂离子电池循环次数的增加出现的老化现象以及锂离子电池性能的下降的问题,本文分析了锂离子电池老化相关参数的变化情况,建立了对应的锂离子电池等效模型并根据实际的电池数据拟合了参数的衰减曲线。同时,基于神经网络自学习的方法设计了针对老化后锂离子电池SOC估算方法。最后,通过数据证明了所提方法能够实现对老化锂离子电池的较为准确的预测。As the number of cycles of a lithium battery increases,the aging phenomenon will arise,along with a de-crease in the battery performance.In this paper,the variations of parameters related to lithium battery aging are ana-lyzed,and a suitable equivalent model for the lithium battery is established.In addition,the decay curve of parameters is fitted according to the actual battery data.Meanwhile,based on the self-learning method of neural network,a state-of-charge(SOC)estimation method for the aging lithium battery is designed.Finally,through data processing,it is proved that the proposed method can realize a more accurate SOC estimation for the aging lithium battery.

关 键 词：动力电池 锂离子电池老化 荷电状态 神经网络 

分 类 号：TM914[电气工程—电力电子与电力传动]