摘要
为了准确估计锂离子电池的健康状态,本文提出一种新的基于改进网格搜索(GS)和广义回归神经网络(GRNN)的估计方法。首先,对集中的数据进行处理,并通过相关性分析方法,提取有效的特征数据,包括电压、电流等。其次,提出一种基于改进网格搜索和广义回归神经网络的回归模型来估计电池的健康状态。最后,使用两个锂离子电池公共数据集验证提出的估计方法。实验结果证明,与其他估计方法相比,所提方法在准确性、泛化性和可靠性方面具有优势。
In order to accurately estimate the state of health(SOH)of lithium-ion batteries,this paper proposes a new estimation method based on improved grid search(GS)and generalized regression neural network(GRNN).Firstly,data processing is performed on the data set,and effective characteristic data including voltage and current are extracted through correlation analysis method.Secondly,a regression model based on improved grid search and generalized regression neural network is proposed to estimate the health of the battery.Finally,two public data sets of lithium-ion batteries are used to verify the proposed estimation method.Experimental results prove that this method has the advantages of accuracy,generalization performance and reliability compared with other estimation methods.
作者
姚远
陈志聪
吴丽君
程树英
林培杰
YAO Yuan;CHEN Zhicong;WU Lijun;CHENG Shuying;LIN Peijie(College of Physics and Information Engineering,Fuzhou University,Fuzhou 350108)
出处
《电气技术》
2021年第7期32-37,共6页
Electrical Engineering
关键词
锂离子电池
健康状态
特征提取
广义回归神经网络(GRNN)
混合算法
lithium-ion battery
state of health(SOH)
feature extraction
generalized regression neural network(GRNN)
hybrid algorithm
