为了提高储能电池健康状态(state of health,SOH)精度及工程适应性,首先,分析某光储电站的运行数据,设计模拟工况实验,结合数据特点和老化机理,选取了3.41 V前30 min内充电电压差(时间间隔分别为1 min、3 min及5 min)作为储能电池健康状态评价的特征参量,并利用20 Ah磷酸铁锂电池的循环数据,结合遗传算法(genetic algorithm,GA)改进的反向传播(back propagation,BP)神经网络进行建模。利用1200次未参与模型训练的数据对模型的精度进行验证,其中1 min间隔电压差为特征参量的模型精度最高,平均绝对百分误差(mean absolute percentage error,MAPE)为0.37%,均方根误差(root mean square error,RMSE)为0.456 5。其次,通过模型迁移,将260 Ah磷酸铁锂电池SOH估计最大误差由5.52%降低到1.89%。最后,利用该模型对光储电站中一簇储能电池单体进行了批量SOH估计,工程适应性良好。
Abstract
关键词
Keywords
references
刘健.基于高斯过程回归的锂离子电池剩余寿命预测研究[D].上海交通大学,2019(06).
Zhang Yu,Zhang Yuhang,Wu Tiezhou.Estimation of state of health based on charging characteristics and back-propagation neural networks with improved atom search optimization algorithm[J].Global Energy Interconnection,2023.
Ma Yan,Shan Ce,Gao Jinwu,Chen Hong.A novel method for state of health estimation of lithium-ion batteries based on improved LSTM and health indicators extraction[J].Energy,2022.
Ezemobi Ethelbert,Tonoli Andrea,Silvagni Mario.Battery State of Health Estimation with Improved Generalization Using Parallel Layer Extreme Learning Machine[J].Energies,2021.
T.R. Ashwin,A. Barai,K. Uddin,L. Somerville,A. McGordon,J. Marco.Prediction of battery storage ageing and solid electrolyte interphase property estimation using an electrochemical model[J].Journal of Power Sources,2018.
Duo Yang,Yujie Wang,Rui Pan,Ruiyang Chen,Zonghai Chen.State-of-health estimation for the lithium-ion battery based on support vector regression[J].Applied Energy,2017.