基于变分贝叶斯无迹卡尔曼滤波SOC估计  

Variational Bayesian-Based Unscented Kalman Filter Method for SOC Estimation

作　　者：汤爱华[1] 刘尚梅 邹航 陈哲明[1] 胡雯爔 李粤涵 TANG Aihua;LIU Shangmei;ZOU Hang;CHEN Zheming;HU Wenxi;LI Yuehan(Key Laboratory of Advanced Manufacturing Technology for Automobile Parts,Ministry of Education,Chongqing University of Technology,Chongqing 400054,China;Information Center of Chongqing University of Technology,Chongqing 400054,China)

机构地区：[1]重庆理工大学汽车零部件先进制造技术教育部重点实验室,重庆400054 [2]重庆理工大学信息中心,重庆400054

出　　处：《西南大学学报（自然科学版）》2024年第12期51-59,共9页Journal of Southwest University(Natural Science Edition)

基　　金：国家自然科学基金项目(52277213);重庆市教委科学技术研究计划重点项目(KJZD-K202201103,KJZD-K202301108)。

摘　　要：锂离子动力电池荷电状态(SOC)的精准估计事关新能源汽车的续驶里程和电池寿命,有助于提高能源利用效率,缓解电池过充与过放等问题.然而,由于过程噪声和观测噪声的存在,锂离子动力电池SOC的估计精度难以保证.鉴于此,提出一种无迹卡尔曼滤波(UKF)和基于变分贝叶斯自适应时变噪声无迹卡尔曼滤波(VBAUKF)联合方法,通过降低过程噪声和观测噪声以实现锂离子动力电池SOC的精确估计.在城市道路循环(UDDS)工况下进行验证,结果表明:锂离子动力电池SOC的估计误差低于1%,验证了所提方法的有效性.Accurate state of charge(SOC)estimation of lithium-ion power battery is related to the range and battery life of new energy vehicles,which is helpful to improve the energy utilization efficiency and alleviate the problems of overcharging and overdischarging of batteries.However,due to the existence of process noise and observation noise,it is hard to guarantee the precision in SOC estimation of lithium-ion power battery.In view of this,this paper proposed a joint method for SOC estimation of lithium-ion power battery by unscented Kalman filtering(UKF)and adaptive time-varying noise unscented Kalman filtering based on variational Bayesian(VBAUKF)to realize the accurate SOC estimation of lithium-ion power battery by reducing the process noise and observation noise.The proposed method was verified under urban dynamometer driving schedule(UDDS)working conditions,and the results showed that the SOC estimation error of lithium-ion power battery was less than 1%,confirming the effectiveness of the proposed method.

关 键 词：锂离子动力电池 SOC估计 无迹卡尔曼滤波 变分贝叶斯 

分 类 号：TK01[动力工程及工程热物理]