A novel health indicator for on-line lithium-ion batteries remaining useful life prediction
A novel health indicator for on-line lithium-ion batteries remaining useful life prediction
复制标题
DOI：
10.1016/j.jpowsour.2016.04.119
复制
发表时间：
2016-07
影响因子：
9.2
通讯作者：
Yapeng Zhou;Miaohua Huang;Yupu Chen;Ye Tao
中科院分区：
文献类型：
--
作者：
Yapeng Zhou;Miaohua Huang;Yupu Chen;Ye Tao
锂离子电池剩余使用寿命预测是智能电池管理系统的重要组成部分。容量和内阻通常用作电池健康指标（HI），用于量化退化和预测RUL。然而，在线测量电容和内阻分别由于未完全充电和放电的条件以及极其昂贵的成本而难以实现。因此，非常需要找到一种可选的方式来应对这种困境。在这项工作中，一个新的HI是从锂离子电池的运行参数中提取的退化建模和RUL预测。此外，Box-Cox变换被用来改善HI性能。然后利用Pearson和斯皮尔曼相关分析来评估真实的容量与由HI得出的估计容量之间的相似性。其次，采用简单的统计回归技术和优化的相关向量机来预测基于所提出的HI的RUL。相关性分析和预测结果表明，所提出的HI电池退化建模和RUL预测的效率和有效性。
Prediction of lithium-ion batteries remaining useful life (RUL) plays an important role in an intelligent battery management system. The capacity and internal resistance are often used as the batteries health indicator (HI) for quantifying degradation and predicting RUL. However, on-line measurement of capacity and internal resistance are hardly realizable due to the not fully charged and discharged condition and the extremely expensive cost, respectively. Therefore, there is a great need to find an optional way to deal with this plight. In this work, a novel HI is extracted from the operating parameters of lithium-ion batteries for degradation modeling and RUL prediction. Moreover, Box-Cox transformation is employed to improve HI performance. Then Pearson and Spearman correlation analyses are utilized to evaluate the similarity between real capacity and the estimated capacity derived from the HI. Next, both simple statistical regression technique and optimized relevance vector machine are employed to predict the RUL based on the presented HI. The correlation analyses and prediction results show the efficiency and effectiveness of the proposed HI for battery degradation modeling and RUL prediction.