基于双向长短期记忆网络含间接健康指标的锂电池SOH估计  

State-of-health Estimation for Lithium-ion Batteries Incorporating Indirect Health Indicators Based on Bi-directional Long Short-term Memory Networks

作　　者：方斯顿 刘龙真 孔赖强 牛涛 陈冠宏 廖瑞金[1] FANG Sidun;LIU Longzhen;KONG Laiqiang;NIU Tao;CHEN Guanhong;LIAO Ruijin(State Key Laboratory of Power Transmission Equipment&System Security and New Technology(Chongqing University),Chongqing 400044,China)

机构地区：[1]输配电装备及系统安全与新技术国家重点实验室(重庆大学),重庆市400044

出　　处：《电力系统自动化》2024年第4期160-168,共9页Automation of Electric Power Systems

基　　金：国家电网公司科技项目(5108-202218280A-2-314-XG)。

摘　　要：快速准确地对锂离子电池进行全寿命周期的健康状态(SOH)估计有助于提高储能设备的安全可靠性。提出一种基于间接健康指标(IHI)和鲸鱼优化算法(WOA)优化的双向长短期记忆(BiLSTM)网络相结合的锂电池SOH估计模型,该模型考虑了未来状态对当前SOH的影响。首先,对锂电池恒流恒压(CC-CV)充放电过程进行分析,提取出多个随充放电循环动态变化的电压、电流、温度的时间特征作为IHI,并加入放电负载电压下降时间这一指标;然后,通过相关性分析,从各IHI中筛选出和容量关联度高的IHI作为输入特征;最后,建立基于WOA优化的BiLSTM网络的电池SOH估计模型,并利用美国国家航天航空局锂电池数据集对2个不同工况下的电池SOH进行估计。结果表明,所提方法可有效提高SOH的估计精度。Rapid and accurate estimation of the state of health(SOH)of lithium-ion batteries throughout their entire life cycle can help improve the safety and reliability of energy storage equipment.An SOH estimation model is proposed,which combines indirect health indicators(IHIs)with bi-directional long short-term memory(BiLSTM)network optimized by the whale optimization algorithm(WOA).The model takes into account the influence of future states on the current SOH.First,the constant current-constant voltage charging and discharging process of lithium-ion battery is analyzed,multiple time characteristics of voltage,current,and temperature that dynamically change with charging and discharging cycles are extracted as IHIs,and the indicator of discharging load voltage drop time is added.Then,through correlation analysis,selected IHIs with high correlation to capacity are set as input features.Finally,a BiLSTM network optimized by WOA is established as the battery SOH estimation model,and the NASA lithium-ion battery dataset is used to estimate the battery SOH under two different operating conditions.The results indicate that the proposed method can effectively improve the estimation accuracy of SOH.

关 键 词：健康状态 锂离子电池 间接健康指标 鲸鱼优化算法 双向长短期记忆网络 

分 类 号：TM912[电气工程—电力电子与电力传动] TP183[自动化与计算机技术—控制理论与控制工程]