重庆理工大学学报2025，Vol.39Issue(7)：17-26,10.DOI:10.3969/j.issn.1674-8425(z).2025.04.003
结合ICA与GS-SVM的电池健康状态估计
State of health estimation of battery based on ICA and GS-SVM
摘要
Abstract
Data-driven approaches are commonly employed for estimating the state of health(SOH)of lithium-ion batteries thanks to their independence from physical models.However,their effectiveness heavily relies on the quality of the characteristic parameters.To enhance the accuracy of battery SOH estimation,this paper proposes a fusion estimation method that combines incremental capacity analysis(ICA)and data-driven approaches.First,the original incremental capacity(IC)curve is smoothed by Gaussian filter,and five characteristic parameters are selected according to the relation between the IC curve and the battery degradation characteristics.Then,the correlation analysis method is employed to identify the top three features that are most strongly correlated with capacity decay as the input of the data-driven model,the support vector machine(SVM)regression prediction model is built to estimate the battery capacity,and the grid search(GS)algorithm is utilized to optimize the SVM parameters.Finally,the effectiveness of the approach is validated by employing publicly available datasets and comparing it with other data-driven methods such as long short-term memory neural network(LSTM),convolutional neural network(CNN),and random forest(RF).Results demonstrate the approach outperforms other data-driven methods in both accuracy and generalization.关键词
锂离子电池/健康状态/增量容量分析/高斯滤波/支持向量机/网格搜索Key words
lithium-ion battery/SOH/ICA/Gaussian filtering/support vector machine/grid search分类
动力与电气工程引用本文复制引用
董静,金帅..结合ICA与GS-SVM的电池健康状态估计[J].重庆理工大学学报,2025,39(7):17-26,10.基金项目
黑龙江省省属高等学校基本科研项目(2023-KYYWF-1013) （2023-KYYWF-1013）
哈尔滨商业大学研究生科研创新项目(YJSCX2023-778HSD) （YJSCX2023-778HSD）
