基于PID型H_(∞)滤波算法估计锂离子电池的SOC  

Estimating SOC of Li-ion battery based on PID controller H_∞ filter

作　　者：吴珞铖 丁洁[1] 姚建鑫 肖敏 WU Luo-cheng;DING Jie;YAO Jian-xin;XIAO Min(School of Automation,School of Artificial Intelligence,Nanjing University of Posts and Telecommunications,Nanjing,Jiangsu 210023,China)

机构地区：[1]南京邮电大学自动化学院,人工智能学院,江苏南京210023

出　　处：《电池》2023年第4期388-392,共5页Battery Bimonthly

基　　金：国家自然科学基金(62073172);江苏省自然科学基金(BK20211275);南京邮电大学自然科学基金(220217,221079)。

摘　　要：针对H_(∞)滤波算法(HIF)对系统状态和模型不确定性突变不敏感的问题,提出一种比例-积分-微分(PID)控制的H_(∞)滤波算法(PID-HIF),以提高荷电状态(SOC)估计的准确性。放电测试中,采用自适应遗忘因子最小二乘(AFFRLS)算法进行系统参数辨识,采用HIF进行SOC估计,与扩展卡尔曼滤波(EKF)算法相比,SOC估计的均方根误差由3.375 3%降低至2.415 8%;在HIF的基础上添加PID控制思想后,均方根误差进一步降低至0.739 3%。在动态应力测试条件下,EKF、HIF和PID-HIF等3种算法的SOC均方根误差分别为4.103 8%、2.414 1%和0.069 4%。Aiming at the issues that the H_∞ filtering algorithm(HIF) was insensitive to sudden changes in system state and model uncertainty,an H_∞ filtering algorithm with proportional integral differential(PID) control(PID-HIF) was proposed to improve the accuracy of SOC estimation.In the discharge test,the adaptive forgetting factor least squares(AFFRLS) algorithm was used to identify the system parameters,HIF was used to estimate SOC.Compared with the extended Kalman filter(EKF) algorithm,the root mean square error of SOC estimation of the method was reduced from 3.375 3% to 2.415 8%,after adding PID control ideas on the basis of HIF,the root mean square error was reduced to 0.739 3%.Under the condition of dynamic stress test,the SOC root mean square error of EKF,HIF and PID-HIF was 4.103 8%,2.414 1% and 0.069 4%,respectively.

关 键 词：锂离子电池 荷电状态(SOC) 在线参数辨识 比例积分微分(PID)的H_(∞)滤波 

分 类 号：TM912.9[电气工程—电力电子与电力传动]