摘要
针对锂离子电池在循环过程中由于充电策略变化引起电池衰退趋势变化从而导致难以追踪的问题,提出了一种适用于不同充电策略的锂电池剩余寿命预测方法。基于NASA与斯坦福-MIT的电池数据提取电池电压平均变化率构建为电池衰减健康因子;将健康因子与电池容量载入基于樽海鞘群算法优化的极限学习机模型之中训练,获取电池在不同充电策略状态下的衰减模型。使用其他充电策略的电池数据对剩余寿命模型进行验证并评估。结果表明:提出的方法能够在确认充电策略情况下准确预测电池寿命,并且在电池运行过程中能够追踪电池容量变化趋势。
Aiming at the problem that the decay trend of the lithium-ion battery changes due to the change of the charging strategy during the cycle,which is difficult to track,a method for predicting the remaining life of the lithium battery suitable for different charging strategies was proposed.Based on the battery data of NASA and Stanford-MIT,the average change rate of battery voltage was extracted to construct the battery decay health factor,and the health factor and battery capacity were loaded into the extreme learning machine model optimized based on the salp swarm algorithm to train to obtain the attenuation model of the battery under different charging strategy states.The remaining life model was verified and evaluated using the battery data of other charging strategies.The experimental results show that the proposed method can accurately predict the battery life under the condition of confirming the charging strategy,and can track the battery capacity change trend during the battery operation.
作者
赵沁峰
蔡艳平
王新军
ZHAO Qinfeng;CAI Yanping;WANG Xinjun(Rocket Force University of Engineering, Xi’an 710025, China)
出处
《兵器装备工程学报》
CSCD
北大核心
2022年第4期250-256,共7页
Journal of Ordnance Equipment Engineering
关键词
锂离子电池
电压平均变化率
樽海鞘群算法
极限学习机
剩余使用寿命
lithium-ion batteries
the average change rate of battery voltage
salp swarm algorithm
extreme learning machine
remaining useful life
