摘要
锂离子电池作为重要的储能元件,其荷电状态(SOC)直接影响所在系统的运行状态。为了实现对锂离子电池SOC的精确估算,提出一种基于门控循环单元神经网络(GRU-RNN)和Huber-M估计鲁棒卡尔曼滤波(HKF)融合方法的锂离子电池SOC估算模型。该方法利用Huber-M估计改进卡尔曼滤波器的鲁棒性,并将基于GRU-RNN所估算的锂离子电池SOC值作为改进卡尔曼滤波器的观测量。在两组锂离子电池数据集上分别进行锂离子电池SOC估算实验。实验结果表明,基于GRU-RNN和HKF融合方法的锂离子电池SOC估算模型不仅能够准确地实现锂离子电池SOC估算,而且能够降低测量误差及异常值对估算结果的影响,使锂离子电池SOC估算结果快速且精确收敛。
As one of the most important energy storage devices,lithium-ion(Li-ion)batteries has been widely used.Accurate and robust state of charge(SOC)estimation of lithium-ion battery is a challenging task in battery management system.In this paper,based on the recurrent neural network with gated recurrent unit(Li-ion),a new hybird model is proposed for SOC estimation.Huber-M estimation is used to improve the robustness of traditional Kalman filter and the output of the GRU-RNN is utilized as the observation of the improved Kalman filter.The performance of proposed methods is evaluated by two experimental datasets.We demonstrate the proposed method achieves satisfactory performance,as well as performs strong robustness against influence of measurement errors and outliers.
作者
李超然
肖飞
樊亚翔
杨国润
唐欣
Li Chaoran;Xiao Fei;Fan Yaxiang;Yang Guorun;Tang Xin(National Key Laboratory of Science and Technology on Vessel Integrated Power System Naval University of Engineering,Wuhan 430033 China)
出处
《电工技术学报》
EI
CSCD
北大核心
2020年第9期2051-2062,共12页
Transactions of China Electrotechnical Society
基金
国防科技创新特区资助项目。
关键词
锂电池
荷电状态
门控循环单元神经网络
卡尔曼滤波
Lithium-ion battery
state of charge
recurrent neural network with gated recurrent unit
Kalman filter
