电动汽车锂离子电池荷电状态估算方法综述  

Review of State of Charge Estimation Methods for Electric Vehicle Lithium-ion Batteries

作　　者：李军[1] 李虎林 LI Jun;LI Hu-lin(School of Mechatronics & Vehicle Engineering, Chongqing Jiaotong University, Chongqing 400074, China)

机构地区：[1]重庆交通大学机电与车辆工程学院,重庆400074

出　　处：《科学技术与工程》2022年第6期2147-2158,共12页Science Technology and Engineering

基　　金：国家自然科学基金(51305472);重庆市研究生联合培养基地(JDLHPYJD2018003)。

摘　　要：电动汽车动力锂电池内部荷电状态估计是电池管理系统状态估计模块的核心,其无法通过仪器直接测量,仅能通过对电池外部电流、电压等参数进行测量并由此估计。准确的荷电状态估计对电池的寿命、容量和安全性管理至关重要。综述了用于电动汽车动力锂电池荷电状态估算的主要方法,根据算法差异将其分为传统的基于传感器测量的开路电压法、电流积分法和阻抗法,基于数据驱动的机器学习类算法以及基于模型的卡尔曼滤波器及粒子滤波器算法与融合类算法。介绍了不同估计算法的计算原理并由此分析比较了不同估计算法的计算复杂度、计算精度等特点。总结了现阶段锂离子电池荷电状态估算研究存在的问题,指出其研究趋势和未来发展方向将是更具泛化性和更高精度以及更佳实时性的多融合类估算方法。The internal state of charge(SOC)estimation of lithium ion battery packs is regarded as the core of the state estimation component to the battery management system(BMS)for the electric vehicles(EVs).But the SOC value can only be estimated now,and the estimation calculation is completed by measuring the value of the external physical parameters of the battery pack,involving the current,voltage and others.The SOC estimation accurately is beneficial for the life extension,energy saving,and safety robustness.The typical calculation ways for the EVs cells-pack SOC estimation were reviewed.These ways can be divided into seven classifications according to the differences in principle:the open circuit voltage(OCV),the current-sensor integration(CSI),the impedance measurement(IM),the artificial neural networks(NNs),the Kalman filters(KF),the particle filters(PF)and the hybrid approaches.The principles of different estimation algorithms were expounded in depth,and the estimation accuracy but complexity of these algorithms were analyzed and compared in detail.Finally,the weakness of research in this field were summarized,and the study trend and orientation in the future were pointed out.The SOC estimation ways will be more generalized,higher-precision and better real-time by reasonable hybrid.

关 键 词：锂离子电池 荷电状态估计 电池管理系统 算法 

分 类 号：TM912.9[电气工程—电力电子与电力传动]