基于多特征融合和CBLS模型的锂离子电池SOH估计方法

江善和; 荣浩威; 张朝龙

doi:10.19799/j.cnki.2095-4239.2026.0639

您当前的位置：

首页 >

文章列表页 >

基于多特征融合和CBLS模型的锂离子电池SOH估计方法

XXXX | 更新时间：2026-09-17

- 基于多特征融合和CBLS模型的锂离子电池SOH估计方法
- Lithium-Ion Battery State-of-Health Estimation Method Based on Multi-Feature Fusion and Convolutional Broad Learning System
- 储能科学与技术 2026年页码：1-18
- 作者机构：
  
  1.安庆师范大学智能制造与机器人学院安庆 246011
  2.金陵科技学院智能科学与控制工程学院南京 211100
- 作者简介：
  
  江善和（1975—），男，博士，教授，基于数据驱动的锂离子电池状态监测，E-mail：jshxlxlw@163.com。
  张朝龙（1982—），男，博士，教授，储能系统运行控制，E-mail：zhangchaolong@126.com。
- 基金信息：
  
  国家自然科学基金项目(51607004);安徽省自然科学基金项目(2008085MF197)
- DOI：10.19799/j.cnki.2095-4239.2026.0639
  中图分类号： TM 614
- 收稿：2026-07-21，
  
  修回：2026-08-25，
  
  网络首发：2026-09-16，
- 稿件说明：
移动端阅览
江善和, 荣浩威, 张朝龙. 基于多特征融合和CBLS模型的锂离子电池SOH估计方法[J]. 储能科学与技术, XXXX, XX(XX): 1-18.

JIANG Shanhe, RONG Haowei, ZHANG Chaolong. Lithium-Ion Battery State-of-Health Estimation Method Based on Multi-Feature Fusion and Convolutional Broad Learning System[J]. Energy Storage Science and Technology, XXXX, XX(XX): 1-18.
江善和, 荣浩威, 张朝龙. 基于多特征融合和CBLS模型的锂离子电池SOH估计方法[J]. 储能科学与技术, XXXX, XX(XX): 1-18. DOI： 10.19799/j.cnki.2095-4239.2026.0639.

JIANG Shanhe, RONG Haowei, ZHANG Chaolong. Lithium-Ion Battery State-of-Health Estimation Method Based on Multi-Feature Fusion and Convolutional Broad Learning System[J]. Energy Storage Science and Technology, XXXX, XX(XX): 1-18. DOI： 10.19799/j.cnki.2095-4239.2026.0639.

摘要

针对锂离子电池健康状态（state of health，SOH）估计中单一健康特征表征能力有限，以及深度学习模型训练复杂、计算开销较大的问题，提出一种基于多特征融合和一维卷积增强宽度学习系统（convolutional broad learning system，CBLS）的SOH估计方法。首先，在电池恒流充电阶段提取循环次数、能量及温度统计特征，采用Pearson相关性分析筛选与SOH具有较强线性关联的循环次数和能量均值，并将表面温度均值作为互补健康特征，构建循环次数、能量均值和温度均值融合的多特征输入。其次，引入一维卷积增强节点改进传统宽度学习系统，实现随机映射特征与卷积增强特征有机融合，充分提取相邻循环局部退化信息，再通过Moore-Penrose广义逆一次性求解输出权重，从而在无须反向传播条件下实现SOH快速估计。力神18650锂离子电池数据集A和B的循环老化数据实验表明，所提的CBLS模型的RMSE值较CNN、BiLSTM和CNN-BiGRU分别降低92.52%和93.67%、83.35%和93.49%、92.93%和92.13%。不同特征组合下的消融实验结果表明，三类特征融合的CBLS在两个数据集上均获得最低估计误差。XJTU公开数据集中23个电池老化数据测试显示，CBLS的平均MAE和RMSE分别为0.178%和0.259%，在不同循环倍率下仍能保持较低的估计误差。所提方法能够有效融合电池老化特征信息，且CBLS模型具有较高的估计精度和泛化性能。

Abstract

To address the limited representation capability of individual health features and the high training complexity and computational cost of deep learning models for lithium-ion battery state-of-health (SOH) estimation

this study proposes an SOH estimation method based on multi-feature fusion and a one-dimensional convolution-enhanced broad learning system (CBLS). Firstly

cycle number and statistical features of charging energy and temperature are extracted from the constant-current charging stage. Pearson correlation analysis is employed to select cycle number and mean charging energy

both of which exhibit strong linear correlations with SOH

while mean surface temperature is introduced as a complementary health feature. Accordingly

a multi-feature input integrating cycle number

mean charging energy

and mean surface temperature is constructed. Secondly

one-dimensional convolutional enhancement nodes are incorporated into the conventional broad learning system to integrate random mapping features with convolutional enhancement features and extract local degradation information across adjacent cycles. The output weights are then analytically determined in a single step using the Moore-Penrose pseudoinverse

enabling rapid SOH estimation without backpropagation. Compared with CNN

BiLSTM

and CNN-BiGRU methods

experiments on cycling-aging Datasets A and B collected from Lishen 18650 lithium-ion batteries show the root mean square error (RMSE)of the proposed CBLS was significantly reduced by 92.52% and 93.67%

83.35% and 93.49%

92.93% and 92.13%

respectively. The ablation experimental results under different feature combinations demonstrate that the proposed CBLS with three types of feature fusion achieves the lowest estimation error on both datasets. Tests on 23 batteries aging data samples from the publicly available XJTU datasets demonstrate that an average mean absolute error (MAE) of the CBLS method are 0.178% and 0.259%

respectively. Also

the CBLS still achieved low estimation errors under different cycling rates. The proposed CBLS method effectively integrates complementary battery degradation information and achieves high SOH estimation accuracy with favorable generalization capability across the evaluated batteries and operating conditions.

关键词

Keywords

references

ELMAHALLAWY M, ELFOULY T, ALOUANI A, et al. A comprehensive review of lithium-ion batteries modeling, and state of health and remaining useful lifetime prediction [J]. IEEE Access, 2022, 10: 119040-119070.

蒋帅, 陈钺, 段砚州, 等. 温度和老化影响下的锂离子电池荷电状态和健康状态联合估计方法研究[J]. 机械工程学报,2024, 60(18): 266-275.

JIANG S, CHEN C, DUAN Y Z, et al. Joint estimation of state-of-charge and state-of-health for lithium-ion batteries under the influence of temperature and aging[J]. Journal of Mechanical Engineering, 2024, 60(18): 266-275.

张朝龙, 刘梦玲, 张俣峰, 等. 基于IEA-T和CNN-BiLSTM-SimAM的锂离子电池健康状态估计[J]. 武汉大学学报(理学版), 2025, 71(3): 385-394.

ZHANG C L, LIU M L, ZHANG Y F, et al. Lithium-ion battery state of health estimation based on IEA-T and CNN-BiLSTM-SimAM [J]. Journal of Wuhan University (Natural Science Edition), 2025, 71(3): 385-394.

张欣怡, 伍嘉浩, 张璐平, 等. 锂离子电池健康状态评估方法研究新进展[J]. 河南科学, 2024, 42(12): 1717-1740.

ZHANG X Y, WU J H, ZHANG L P, et al. New research progress of the state of health estimation methods for lithium-ion batteries[J]. Henan Science, 2024, 42(12): 1717-1740.

WANG Y P, KIM S, CHEN Y F, et al. An SMPS-based lithium-ion battery test system for internal resistance measurement[J]. IEEE Transactions on Transportation Electrification, 2023, 9(1): 934-944.

PARK J, SON H, GEUM S, et al. Effect of terminal resistance and SOH asymmetry on parallel cell interaction and battery system degradation[J]. Journal of Energy Storage, 2025, 137: 118536.

LI C F, YANG L, LI Q S, et al. SOH estimation method for lithium-ion batteries based on an improved equivalent circuit model via electrochemical impedance spectroscopy[J]. Journal of Energy Storage, 2024, 86: 111167.

李清波, 张懋慧, 罗英, 等. 基于等效电路模型融合电化学原理的锂离子电池荷电状态估计[J]. 储能科学与技术, 2024, 13(9): 3072-3083.

LI Q B, ZHANG M H, LUO Y, et al. Lithium-ion battery state of charge estimation based on equivalent circuit model[J]. Energy Storage Science and Technology, 2024, 13(9): 3072-3083.

GAO Y Z, LIU K, ZHU C, et al. Co-estimation of state- of-charge and state-of-health for lithium-ion batteries using an enhanced electrochemical model[J]. IEEE Transactions on Industrial Electronics, 2022, 69(3): 2684–2696.

BAGHDADI I, BRIAT O, DELETAGE J, et al. Lithium battery aging model based on Dakin's degradation approach[J]. Journal of Power Sources, 2016, 325: 273-285.

巫春玲,王立顶,卢勇,等.基于白鹭群优化高斯过程回归的锂电池SOH估计方法[J]. 储能科学与技术, 2025, 14(6) : 2498-2511.

WU C L, WANG L D, LU Y, et al. Lithium-ion batteries SOH estimation based on gaussian processed regression optimized by egret swarm optimization[J]. Energy Storage Science and Technology, 2025, 14(6): 2498-2511.

毛百海, 覃吴, 肖显斌, 等. 基于LSTM&GRU-Attention多联合模型的锂离子电池SOH估计[J]. 储能科学与技术, 2023, 12(11): 3519-3527.

MAO B H, QIN W, XIAO X B, et al. SOH estimation of lithium-ion batteries based on LSTM&GRU-Attention multijoint model[J]. Energy Storage Science and Technology, 2023, 12(11): 3519-3527.

张宣梁, 何霆, 朱文龙, 等. 基于多循环特征的储能电池SOH估计模型[J]. 储能科学与技术, 2023, 12(11): 3488-3498.

ZHANG X L , HE T , ZHU W L, et al. A SOH estimation model for energy storage batteries based on multiple cycle features[J]. Energy Storage Science and Technology, 2023, 12(11): 3488-3498.

XIONG W, MO Y M, YAN C. Online state-of-health estimation for second-use lithium-ion batteries based on weighted least squares support vector machine[J]. IEEE Access, 2021, 9: 1870-1881.

车云弘, 邓忠伟, 李佳承, 等. 基于数据驱动的电池系统泛化SOH估计方法[J]. 机械工程学报, 2022, 58(24): 253-263.

CHE Y H, DENG Z W, LI J C, et al. Generalized data-driven SOH estimation method for battery systems[J]. Journal of Mechanical Engineering, 2022, 58(24): 253-263.

郭鹏旭, 赵理, 张丰硕. 基于随机充电片段的锂电池健康状态估计方法[J ] . 电源学报, 2024, 1 - 13 .

GUO P X, ZHAO L, ZHANG F S. A method for estimating the State of health of lithium-ion batteries based on random charging segments[J]. Journal of Power Supply, 2024, 1-13.

刘富强, 刘为国, 朱洪波. 基于灰狼算法优化Transformer-BiLSTM的锂电池健康状态预测方法[J ] . 重庆工商大学学报(自然科学版), 2025, 1 - 12 .

LIU F Q, LIU W G, ZHU H B. Optimized Transformer-BiLSTM based on gray wolf algorithm for health state prediction of lithium battery[J]. Journal of Chongqing Technology and Business University (Natural Science Edition), 2025, 1-12.

时洪雷, 刘喜军, 高竟译, 等. 基于超参数优化和混合神经网络的锂硫电池健康状态估计[J ] . 储能科学与技术, 2025, 1 - 11 .

SHI H L, LIU X J, GAO J Y, et al. State of health estimation of lithium-sulfur batteries based on hyperparameter optimization and hybrid neural networks[J]. Energy Storage Science and Technology, 2025, 1-11.

武骥, 方雷超, 刘兴涛, 等. 基于特征优化和随机森林算法的锂离子电池SOH估计[J]. 机械工程学报, 2024, 60(12): 335-343.

WU J, FANG L C, LIU X T, et al. State of health estimation of lithium-ion battery based on feature optimization and random forest algorithm[J]. Journal of Mechanical Engineering, 2024, 60(12): 335-343.

陆继忠, 彭思敏, 李晓宇. 基于多特征量分析和LSTM-XGBoost模型的锂离子电池SOH估计方法[J]. 储能科学与技术, 2024, 13(9): 2972-2982.

LU J Z, PENG S M, LI X Y. State-of-health estimation of lithium-ion batteries based on multifeature analysis and LSTM-XGBoost model[J]. Energy Storage Science and Technology, 2024, 13(9): 2972-2982.

耿陈, 孟锦豪, 彭乔, 等. 基于弛豫过程特征提取的锂离子电池健康状态估计[J]. 储能科学与技术, 2023, 12(11): 3479-3487.

GENG C, MENG J H, PENG Q, et al. Estimation of the state of health of lithium-ion batteries based on feature extraction of the relaxation process[J]. Energy Storage Science and Technology, 2023, 12(11): 3479-3487.

许培德, 刘康, 康龙云, 等. 基于弛豫电压和BO-DNN的锂离子电池健康状态估计[J ] . 电源学报, 2024, 1 - 14 .

XU P D, LIU K, KANG L Y, et al. State of health estimation of lithium-ion battery based on relaxation voltage and BO-DNN[J]. Journal of Power Supply, 2024, 1-14.

郭鹏旭, 赵理, 马占潮, 等. GWO-CNN框架下的锂电池健康状态估计方法[J]. 重庆理工大学学报(自然科学), 2025, 39(4): 9-16.

GUO P X, ZHAO L, MA Z C, et al. Lithium battery health status estimation method under GWO-CNN framework[J]. Journal of Chongqing University of Technology ( Natural Science), 2025, 39(4): 9-16.

CAI L, CUI N M, JIN H Y, et al. A unified deep learning optimization paradigm for lithium-ion battery state-of-health estimation[J]. IEEE Transactions on Energy Conversion, 2024, 39(1): 589-600.

ZHANG C L, LUO L J, YANG Z, et al. Battery SOH estimation method based on gradual decreasing current,double correlation analysis and GRU[J]. Green Energy and Intelligent Transportation, 2023, 2: 100108.

GUI X A, DU J R, SONG L, et al. A Transformer-CNN based transferable model for state-of-health prediction of lithium-ion batteries[C]. 2023 Global Reliability and Prognostics and Health Management Conference (PHM-Hangzhou), Hangzhou, China, 2023, 1-8.

SV M, D R, SUTHA S, et al. Data-driven model for estimation of SOH and internal resistance of lithium-ion battery based on deep neural network[C]. 2024 Tenth Indian Control Conference (ICC), Bhopal, India, 2024, 446-451.

YU Z W, PAN Z N. Joint estimation of SOH and RUL for lithium-ion batteries using CNN-GRU based on artificial lemming algorithm[C]. 2025 IEEE 3rd International Conference on Power Science and Technology (ICPST), Kunming, China, 2025, 1269-1274.

YANG S J, ZHANG C P, JIANG J C, et al. Review on state-of-health of lithium-ion batteries: Characterizations, estimations and applications[J]. Journal of Cleaner Production, 2021, 314: 128015.

CHEN C L P, LIU Z. Broad learning system: an effective and efficient incremental learning system without the need for deep architecture[J]. IEEE Transactions on Neural Networks and Learning Systems, 2017, 29(1): 10-24.

ZHAO J J, ZHANG X Y, HU C H. Lithium-ion battery state-of-health estimation using voltage-position encoding CNN and incremental capacity analysis with a novel smoothing parameter selection strategy[J]. Journal of Energy Storage, 2025, 130: 117296.

LI S H, JIANG Z L, ZHU Z W, et al. A framework of joint SOC and SOH estimation for lithium-ion batteries: using BiLSTM as a battery model[J]. Journal of Power Sources, 2025, 635: 236342.

SHI Z L, XU J Z, IBRAHIM A, et al. Accurate state of health estimation based on the multi-feature extracted and improved MHA assisted CNN-BiGRU model for lithium-ion batteries[J]. Journal of Energy Storage, 2025, 131: 117598.

Wang F, Zhai Z, Zhao Z, et al. Physics-informed neural network for lithium-ion battery degradation stable modeling and prognosis[J]. Nature Communications, 2024, 15(1): 4332.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

一种并行多尺度特征融合模型开展的基于弛豫电压的锂电池SOH估计研究

锂离子电池组分层均衡控制系统研究

基于弛豫电压低频特征的锂电池容量在线评估方法

面向快充的锂离子电池多尺度三维导电网络结构设计与离子传输优化

层状/橄榄石混合正极材料的研究进展：从机制协同到性能增强