1、基于扩展卡尔曼滤波和神经网络组合模型的电池荷电状态的测定Zhihang Chena, Shiqi Qiua, M.Abul Masrurb, Yi Lu MurpheyaaDepartment of Electrical and Computer Engineering, The University of Michigan-Dearborn,Dearborn, MI 48128, USAbU.S. Army RDECOM-TARDE, Warren, MI, USA【摘要】本文主要介绍智能电池管理系统中电池荷电状态（SOC）的估测。研究的重点是使用卡尔曼滤波器和神经网络组合模型去估测动态 S
2、OC。首先，我们研发了一种利用扩展型卡尔曼滤波器（EKF）去模拟电池滞后影响的方法。其次，我们设计了一种 SOC 估测模型，NN-EKF 模型，这种模型用 EKF 嵌入神经网络组成。这个假设方法已经利用两种不同电池锂离子电池 U1-12XP 和 1.2V，3.4Ah 的镍氢电池的真实数据得到了验证。我们的试验显示了和其他先进的试验方法相比，用 EKF 模型模拟基于独立充放电的开路电压（OCV）曲线电池滞后现象，在 SOC 估测上有着最佳的表现。其次，NN-EKF 模型在没有或者有温度数据的情况下估测 SOC 都有着最好的表现。关键词电池荷电状态（SOC） ；卡尔曼滤波器，神经网络，智能电池管理
3、策略 介绍为了响应人们对环境保护和能源保护问题的日益关切，对于混合电动汽车和纯电动汽车（HEVs C. Cho, “ Estimation of the stateof charge of Ni-MH battery pack based on artificial neural network”,INTELEC 2009. 31st International Telecommunications EnergyConference, 2009.4 G. L. Plett, “Battery management system algorithms for HEV batterystate-o
4、f-charge and state-of-health estimation”, Transworld ResearchNetwork5 S. Piller, M. Perrin, A. Jossen, “Methods for state-of-chargedetermination and their applications”, Journal of Power Sources 96(2001) 113-1206 G. L. Plett, “Extended Kalman filtering for battery managementsystems of LiPB-based HEV
5、 battery packs: Part 1. Background”, J.Power Sources 134 25261, 20047 G. L. Plett, “Extended Kalman filtering for battery managementsystems of LiPB-based HEV battery packs: Part 2. Modeling andidentification”, J. Power Sources 134 26276, 20048 G. L. Plett, “Extended Kalman filtering for battery mana
6、gementsystems of LiPB-based HEV battery packs: Part 3. State andparameter estimation”, J. Power Sources 134 27792, 20049 J. Wang, B. Cao, Q. Chen, F. Wang, Combined state of chargeestimator for electric vehicle battery pack, Control EngineeringPractice 15 (2007) 15691576.10 J. Wang, J. Guo, L. Ding,
7、 An adaptive Kalman filtering based State ofCharge combined estimator for electric vehicle battery pack, EnergyConversion and Management 50 (2009) 3182318611 A. Vasebi , M. Partovibakhsh, S. Mohammad Taghi Bathaee, A novelcombined battery model for state-of-charge estimation in lead-acidbatteries ba
8、sed on extended Kalman filter for hybrid electric vehicleapplications, Journal of Power Sources 174 (2007) 304012 K. S. Ng, C. Moo, Y. Chen, Y. Hsieh, Enhanced coulomb countingmethod for estimating state-of-charge and state-of-health of lithiumionbatteries, Applied Energy 86 (2009) 1506151113 http:/ http:/www.cs.unc.edu/welch/kalman/
