荷电状态
残余物
估计员
递归最小平方滤波器
控制理论(社会学)
协方差矩阵
卡尔曼滤波器
算法
扩展卡尔曼滤波器
稳健性(进化)
计算机科学
数学
数学优化
电池(电)
自适应滤波器
统计
人工智能
化学
功率(物理)
物理
生物化学
控制(管理)
量子力学
基因
作者
Xugang Zhang,Linchao Duan,Qingshan Gong,Yan Wang,Huawei Song
标识
DOI:10.1016/j.jpowsour.2023.233758
摘要
The accurate estimation of the state of charge (SOC) in lithium-ion batteries plays a pivotal role in battery management systems. This paper proposes an improved adaptive extended Kalman filter (IAEKF) algorithm for more precise SOC estimation. Initially, a novel characteristic parameter is introduced to assess the suitability of the forgetting factor in the adaptive forgetting factor recursive least squares (AFFRLS) algorithm. This evaluation aims to improve the accuracy of recognizing the parameters of the battery model. Additionally, an even function with adjustable parameters is constructed to solve the appropriate value for the forgetting factor. Subsequently, the correlation coefficients of the residual series at different moments are approximated by mathematical transformations that link the inner products of the column vectors in the constructed residual matrix to the calculation of the correlation coefficients. By comparing it with optimized thresholds using particle swarm optimization, one can adjust the sliding window length to enhance the estimation accuracy of IAEKF. The experimental results confirm that IAEKF achieves superior estimation accuracy and robustness compared to other AEKF algorithms, and the accuracy of AFFRLS parameter identification is higher than that of recursive least squares, thereby substantiating the effectiveness of the proposed algorithms.
科研通智能强力驱动
Strongly Powered by AbleSci AI