基于二层分解技术的锂离子电池容量评估方法

Capacity estimation method of lithium-ion batteries based on two-layer decomposition technique

锂离子电池健康状态(SOH)能够表征当前电池老化程度,分析当前各类SOH估算方法,存在无法直接测量及确定数量合适的估算输入量等问题。为了解决这些问题,从容量角度定义SOH,选择可以在线测量的等压降放电时间作为健康因子,构建改进快速集合经验模态分解(FEEMD)与可变模式分解(VMD)相结合的二层分解技术和粒子群算法优化长短期记忆网络(LSTM)实现锂离子电池容量估算,获得SOH值。基于NASA锂离子电池数据集进行实验。结果表明,利用该方法能够有效地对锂电池的健康趋势进行拟合,得到准确的SOH估算结果。

The health of state(SOH)of lithium-ion batteries can represent the current battery aging degree.By analyzing the various SOH estimation methods,the problems of inability to directly measure SOH and determine the amount of estimation input have not been solved.In order to solve these problems,SOH was defined with capacity.The constant voltage discharge time was chosen as health factors.The fast ensemble empirical mode decomposition(FEEMD)was combined with variable mode decomposition(VMD)to construct two-layer decomposition technique.The particle swarm optimization was used for optimizing long short-term memory network(LSTM)to estimate the lithium-ion battery capacity,obtaining SOH.The experiment was conducted by using NASA's lithium-ion battery data set.The results show that the proposed method can effectively fit the health trend of lithium batteries and obtain accurate SOH estimation results.