A Novel Framework for the SOC and SOH Co-estimation Based on AEKF and LSTM
作者单位：School of Mechanical EngineeringHebei University of Technology State Key Laboratory of Reliability and Intelligence of Electrical EquipmentHebei University of Technology
会议名称：《2023中国汽车工程学会年会暨展览会》
会议日期：2023年
学科分类：082304[工学-载运工具运用工程] 08[工学] 080204[工学-车辆工程] 0802[工学-机械工程] 0823[工学-交通运输工程]
摘 要：With new energy vehicles gradually dominate the market,more and more attentions are paid to the safe operation and long-life cycle of battery systems,and the accurate estimation of state of charge(SOC) and state of health(SOH) are the key functions of *** this study,a novel approach for simultaneously estimating the SOC and SOH is *** estimation methods for SOC and SOH are fused the model-based and data-driven ***,an adaptive extended Kalman filter(AEKF) is applied to estimate SOC without requiring the SOH estimate update parameter,and long short-term memomy(LSTM) is utilized as the data-driven method for SOH *** the data-driven method,the health features(HFs) are extracted from the incremental capacity(IC)curves and SOC-OCV curves,and in this study,six kinds of health factors based on different requirements are ***,the proposed method is validated using a dataset with the CACLE dataset with complex conditions and the Oxford dataset,and the results indicate that the errors of SOC estimation are less than 1% and the errors of SOH are within 3.8%.The results show that the proposed framework has practical value.