首页|基于潜变量强跟踪滤波的锂离子电池荷电状态估计

基于潜变量强跟踪滤波的锂离子电池荷电状态估计

扫码查看
文章提出一种基于潜变量强跟踪滤波算法的锂电池荷电状态估计方法.该方法将非线性部分的基本函数项设定为潜在变量,并将系统的初始变量和潜在变量相结合,从而将初始的系统模型提升到高维的线性状态模型.该方法通过将非线性部分的基本函数项定义为潜变量,建立潜变量与原始变量之间的线性模型,避免了泰勒展开,同时引入渐消因子来补偿因动态关联模型带来的误差.通过马里兰大学实验室数据在电池动态应力测试和美国联邦城市运行工况下进行仿真实验,验证了新方法在锂电池荷电状态估计的精确度和适用性.实验结果表明,潜变量强跟踪滤波算法相较于扩展卡尔曼滤波算法估计误差更小,精确度更高.
State of Charge Estimation of Lithium-ion Batteries Based on Latent Variable Strong Tracking Filtering
A state of charge(SOC)estimation method for lithium batteries based on latent variable strong tracking filtering algorithm(HV-STF)is proposed.This method sets the basic function terms of the nonlinear part as latent variables and combines the initial and latent variables of the system,thereby elevating the initial system model to a high-dimensional linear state model.In addition,the observation model is also transformed into a high-dimensional linear state observation model through equivalent rewriting.Through this approach,the latent variable strong tracking filtering algorithm can more accurately estimate the SOC value of lithium batteries.After experimental verification,this method has shown high accuracy in estimating the state of charge(SOC)of lithium batteries.Compared with Extended Kalman Filter(EKF),this method can more accurately estimate the SOC of lithium batteries,which is of great significance for improving the operational safety and performance of lithium batteries.

latent variablesstate of chargelithium batteriesKalman filtering

史永辉、文成林

展开 >

吉林化工学院 信息与控制工程学院,吉林 132000

广东石油化工学院 自动化学院,广东 茂名 525000

潜变量 荷电状态 锂电池 卡尔曼滤波

2024

广东石油化工学院学报
广东石油化工学院

广东石油化工学院学报

影响因子:0.2
ISSN:2095-2562
年,卷(期):2024.34(4)