A distribution-free interval estimation method for lithium-ion battery state of health

IF 2.6 4区化学 Q3 CHEMISTRY, PHYSICAL

Ionics Pub Date : 2025-06-17 DOI:10.1007/s11581-025-06470-3

Xiaoqiong Pang, Ziyao Guo, Jianfang Jia, Jie Wen, Xiaojie Li, Jianchao Zeng, Jiashuo Zhang

{"title":"A distribution-free interval estimation method for lithium-ion battery state of health","authors":"Xiaoqiong Pang, Ziyao Guo, Jianfang Jia, Jie Wen, Xiaojie Li, Jianchao Zeng, Jiashuo Zhang","doi":"10.1007/s11581-025-06470-3","DOIUrl":null,"url":null,"abstract":"<div><p>In order to estimate the state of health (SOH) of lithium-ion batteries, traditional methods are often based on distributional assumptions that may not match the actual situation and lead to the construction of prediction intervals (PIs) that are unreliable. To this end, a distribution-free interval estimation strategy for SOH in lithium-ion batteries is proposed in this paper, which aims to efficiently construct high-quality PIs. To ensure compatibility with the gradient descent (GD) algorithm and to overcome the previous reliance on meta-heuristics, the loss function is redesigned and the PI centre is innovatively incorporated into the optimization objective. The aim is to obtain high-quality PIs for a more comprehensive assessment of their quality. Through the validation of the NASA dataset and the CALCE dataset, the results show that the proposed method improves the comprehensive evaluation metric <i>P</i> by an average of 40.69% compared to the traditional lower upper bound estimation (LUBE) method and improves the mean PI centre deviation (MPICD) metric by 13.60% compared to the model that does not consider the PI centre.</p></div>","PeriodicalId":599,"journal":{"name":"Ionics","volume":"31 8","pages":"7939 - 7952"},"PeriodicalIF":2.6000,"publicationDate":"2025-06-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Ionics","FirstCategoryId":"92","ListUrlMain":"https://link.springer.com/article/10.1007/s11581-025-06470-3","RegionNum":4,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"CHEMISTRY, PHYSICAL","Score":null,"Total":0}

引用次数: 0

Abstract

In order to estimate the state of health (SOH) of lithium-ion batteries, traditional methods are often based on distributional assumptions that may not match the actual situation and lead to the construction of prediction intervals (PIs) that are unreliable. To this end, a distribution-free interval estimation strategy for SOH in lithium-ion batteries is proposed in this paper, which aims to efficiently construct high-quality PIs. To ensure compatibility with the gradient descent (GD) algorithm and to overcome the previous reliance on meta-heuristics, the loss function is redesigned and the PI centre is innovatively incorporated into the optimization objective. The aim is to obtain high-quality PIs for a more comprehensive assessment of their quality. Through the validation of the NASA dataset and the CALCE dataset, the results show that the proposed method improves the comprehensive evaluation metric P by an average of 40.69% compared to the traditional lower upper bound estimation (LUBE) method and improves the mean PI centre deviation (MPICD) metric by 13.60% compared to the model that does not consider the PI centre.

查看原文本刊更多论文

锂离子电池健康状态的无分布区间估计方法

为了估计锂离子电池的健康状态，传统的方法往往基于分布假设，可能与实际情况不匹配，导致预测区间（pi）的构建不可靠。为此，本文提出了一种锂离子电池SOH无分布区间估计策略，旨在高效构建高质量的pi。为了确保与梯度下降（GD）算法的兼容性并克服先前对元启发式算法的依赖，对损失函数进行了重新设计，并创新地将PI中心纳入优化目标。目的是获得高质量的pi，以便对其质量进行更全面的评估。通过对NASA数据集和CALCE数据集的验证，结果表明，所提方法与传统上界估计（LUBE）方法相比，综合评价指标P平均提高了40.69%，与不考虑PI中心的模型相比，平均PI中心偏差（MPICD）指标提高了13.60%。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Ionics 化学-电化学

CiteScore

5.30

自引率

7.10%

发文量

427

审稿时长

2.2 months

期刊介绍： Ionics is publishing original results in the fields of science and technology of ionic motion. This includes theoretical, experimental and practical work on electrolytes, electrode, ionic/electronic interfaces, ionic transport aspects of corrosion, galvanic cells, e.g. for thermodynamic and kinetic studies, batteries, fuel cells, sensors and electrochromics. Fast solid ionic conductors are presently providing new opportunities in view of several advantages, in addition to conventional liquid electrolytes.