混合模糊信息粒化和时间序列密集编码器的锂离子电池剩余寿命区间预测方法

Hybrid fuzzy information granulation and time series dense encoder method for interval prediction of lithium-ion battery remaining useful life

锂离子电池的剩余使用寿命(remaining useful life,RUL)是电池健康状态的关键指标,对其进行预测具有重要的现实意义。该工作将模糊信息粒化(fuzzy information granulation,FIG)技术与时间序列密集编码器模型(timeseries dense encoder,TiDE)相结合,提出了一种对锂离子电池的RUL进行区间预测的模型。首先将锂离子电池容量退化时间序列通过FIG技术转化为粒子序列信息,以此得到模糊信息粒子的上下界序列。其次,分别对上下界序列使用TiDE模型进行训练预测,从而得到区间预测的结果。实验结果表明,与基于支持向量回归(support vector regression,SVR)和长短期记忆网络(long short term memory network,LSTM)的区间预测模型以及不使用狐狸优化算法(fox-inspired optimization algorithm,FOA)优化的TiDE模型相比,该工作提出的基于FIG技术结合TiDE模型与FOA的区间预测方法在锂离子电池RUL预测性能上具有更高的可靠性。

The remaining useful life(RUL)of lithium-ion battery is a critical indicator of battery health,and accurately predicting it holds significant practical importance.This work combines fuzzy information granulation(FIG)technology with the time-series dense encoder(TiDE)model,to propose an interval prediction model for the RUL of lithium-ion battery.First,the capacity degradation time series of lithium-ion battery is transformed into particle sequence information using FIG technology,resulting in the upper and lower bound sequences of fuzzy information particles.Then,the TiDE model is trained and used to predict the upper and lower bound sequences separately,yielding interval prediction results.Experimental results indicate that,compared to interval prediction models based on support vector regression(SVR)and long short term memory network(LSTM)and the TiDE model without fox-inspired optimization algorithm(FOA),the proposed interval prediction method,which combines FIG technology with the TiDE model and FOA,offers higher reliability in predicting the RUL of lithium-ion battery.