An algorithm-hybrid observer combining proportional-integral with Kalman filter for state-of-charge estimation of lithium-ion battery  

作　　者：Guangwei YIN Hua WANG Lin HE Xiaofei LIU Guoqiang WANG Jichao LIU 

机构地区：[1]School of Automotive and Transportation Engineering,Hefei University of Technology,Hefei,230009,China [2]Laboratory of Automotive Intelligence and Electrification,Hefei University of Technology,Hefei,230009,China [3]College of Foreign Languages,Shandong University of Science and Technology,Qingdao,266590,China [4]XCMG Construction Machinery Research Institute,Xuzhou,221004,China

出　　处：《Science China(Technological Sciences)》2025年第3期298-309,共12页中国科学(技术科学英文版)

基　　金：supported by the Key Research and Development Program of Jiangsu Province(Grant No.BE2021006-2);the Key Science and Technology Program of Anhui Province(Grant No.202423d12050001);the Natural Science Foundation of Anhui Province(Grant No.2308085ME163);the National Natural Science Foundation of China(Grant No.62103415)。

摘　　要：Estimating the state-of-charge(SOC)of lithium-ion batteries faces three main challenges at present:ensuring accuracy,achieving smooth output,and maintaining low computational complexity.To tackle these issues,this study introduces a hybrid algorithm observer.This approach combines the proportional-integral(PI)principle with the Kalman filter,utilizing a state-of-charge dynamics model and a current dynamics model.The SOC dynamics model,described by a differential equation,is developed to improve estimation accuracy.Meanwhile,the current dynamics model supports the design of a PI observer,which offers a low-complexity solution for SOC estimation.To address the issue of white noise in measurement signals,a onedimensional Kalman filter is applied.This filter smooths the output signal and enhances accuracy by addressing the limitations of the PI observer.In addition,the system incorporates parameter observation to estimate key battery parameters.The hybrid observer was tested in a real vehicle to validate its effectiveness.Experimental results and statistical analysis demonstrate that this algorithm is a strong candidate for accurately estimating SOC in lithium-ion batteries.

关 键 词：battery charge characteristics current-integral method battery parameters observation current dynamics model state-of-charge dynamics model 

分 类 号：TM912[电气工程—电力电子与电力传动]