Modified Gaussian Process Regression Models for Cyclic Capacity Prediction of Lithium-Ion Batteries
Modified Gaussian Process Regression Models for Cyclic Capacity Prediction of Lithium-Ion Batteries
复制标题
锂离子电池循环容量预测的修正高斯过程回归模型
DOI：
10.1109/tte.2019.2944802
复制
发表时间：
2019-12-01
影响因子：
7
通讯作者：
Jiang, Yan
中科院分区：
文献类型：
--
作者：
Liu, Kailong;Hu, Xiaosong;Jiang, Yan
本文介绍了支持机器学习的数据驱动模型的开发，用于预测不同循环条件下锂离子 (Li-ion) 电池的有效容量。为了实现这一目标，首先提出了考虑电池老化趋势以及相应的工作温度和放电深度的模型结构。然后，基于对高斯过程回归 (GPR) 中协方差函数的系统理解，开发了两个相关的数据驱动模型。具体来说，通过使用自动相关性确定结构修改各向同性平方指数核，“模型A”可以提取高度相关的输入特征以进行容量预测。通过将阿累尼乌斯定律和多项式方程耦合到组合核中，“模型 B”能够考虑电池退化的电化学和经验知识。开发的模型在具有各种循环模式的镍锰钴（NMC）氧化物锂离子电池上进行了验证和比较。实验结果表明，考虑电池电化学和经验老化特征的改进探地雷达模型优于其他模型，并且能够在一步和多步预测方面取得令人满意的结果。所提出的技术有望用于各种循环情况下的电池容量预测。
This article presents the development of machine-learning-enabled data-driven models for effective capacity predictions for lithium-ion (Li-ion) batteries under different cyclic conditions. To achieve this, a model structure is first proposed with the considerations of battery aging tendency and the corresponding operational temperature and depth-of-discharge. Then based on a systematic understanding of the covariance functions within the Gaussian process regression (GPR), two related data-driven models are developed. Specifically, by modifying the isotropic squared exponential kernel with an automatic relevance determination structure, “Model A” could extract the highly relevant input features for capacity predictions. Through coupling the Arrhenius law and a polynomial equation into a compositional kernel, “Model B” is capable of considering the electrochemical and empirical knowledge of battery degradation. The developed models are validated and compared on the nickel–manganese–cobalt (NMC) oxide Li-ion batteries with various cycling patterns. The experimental results demonstrate that the modified GPR model considering the battery electrochemical and empirical aging signature outperforms other counterparts and is able to achieve satisfactory results for both one-step and multistep predictions. The proposed technique is promising for battery capacity predictions under various cycling cases.