基于 CNN-LSTM 融合模型的动力电池
健康状态精准评价研究

doi:10.16638/j.cnki.1671-7988.2025.019.006

汽车实用技术 ›› 2025, Vol. 50 ›› Issue (19): 31-36.DOI: 10.16638/j.cnki.1671-7988.2025.019.006

• 新能源汽车 • 上一篇

基于 CNN-LSTM 融合模型的动力电池健康状态精准评价研究

柯胜蓝，李丹*，吕亚帆

上汽大众汽车有限公司

发布日期:2025-10-09
通讯作者: 李丹
作者简介:柯胜蓝（1975－），男，工程师，硕士，研究方向为汽车材料开发与应用通信作者：李丹（1990－），男，工程师，硕士，研究方向为汽车材料开发与应用

Research on Accurate Evaluation of Power Battery Health State Based on CNN-LSTM Hybrid Model

KE Shenglan, LI Dan* , LV Yafan

SAIC Volkswagen Automotive Company Limited

Published:2025-10-09
Contact: LI Dan

摘要/Abstract

摘要： 动力电池梯次利用依赖精准的电池健康状态（SOH）评价，但现有方法存在依赖单一特征、忽略时序动态变化等问题。文章提出一种基于工业计算机断层扫描（CT）影像的卷积神经网络-长短期记忆网络（CNN-LSTM）融合模型，通过工业 CT 影像提取电芯内部三维结构特征，结合 CNN-LSTM 融合模型同步建模空间特征与时序退化规律，实现 SOH 的精准预测。实验表明，该方法的均方误差（MSE）为 0.033，较传统内阻法和电化学阻抗谱（EIS）法，平均绝对误差（MAE）低超 30%（从 0.038 降至 0.026），且在不同 SOH 区间（40%～100%）的预测误差率均低于 4.2%。该研究为梯次利用电池筛选提供了高精度工具，并推动新能源产业的可持续发展。

关键词: 梯次利用；电池健康状态评价；卷积神经网络；长短期记忆网络；工业 CT

Abstract: The graded utilization of power batteries relies on accurate state of health (SOH) evaluation. However, existing methods suffer from limitations such as dependence on single features and neglect of temporal dynamic changes. This paper proposes a hybrid convolutional neural network-long short-term memory (CNN-LSTM) model based on industrial computed tomography (CT) images. The method extracts three-dimensional internal structural features of battery cells from industrial CT images, and the CNN-LSTM hybrid model simultaneously captures spatial features and temporal degradation patterns to achieve accurate SOH prediction. Experimental results demonstrate that the proposed method achieves a mean squared error (MSE) of 0.033. Compared to traditional internal resistance methods and electrochemical impedance spectroscopy (EIS) methods, the mean absolute error (MAE) is reduced by over 30% (from 0.038 to 0.026). Moreover, the prediction error rate remains below 4.2% across different SOH ranges (40%～100%). This study provides a highprecision tool for screening batteries in graded utilization and promotes the sustainable development of the new energy industry.

Key words: graded utilization; battery state of health evaluation; convolutional neural network; long short-term memory; industrial CT

柯胜蓝，李丹*，吕亚帆. 基于 CNN-LSTM 融合模型的动力电池健康状态精准评价研究[J]. 汽车实用技术, 2025, 50(19): 31-36.

KE Shenglan, LI Dan* , LV Yafan. Research on Accurate Evaluation of Power Battery Health State Based on CNN-LSTM Hybrid Model[J]. Automobile Applied Technology, 2025, 50(19): 31-36.

参考文献

[ 1 ] 中国动力电池回收利用产业联盟.2023 年中国动力电池回收利用行业发展报告[R].北京:中国动力电池回收利用产业联盟,2024. [ 2 ] 工业和信息化部科技部生态环境部商务部市场监管总局.关于印发《新能源汽车动力蓄电池梯次利用管理办法》的通知[EB/OL].(2021-08-19)[2024- 10-10].https://www.gov.cn/zhengce/zhengceku/2021- 08/28/content_5633897.htm. [ 3 ] 国家发展改革委国家能源局.关于加快推动新型储能发展的指导意见[EB/OL].(2021-07-15)[2024-10- 10].http://zfxxgk.nea.gov.cn/2021-07/15/c_13100793 31.htm. [ 4 ] 赵珈卉,田立亭,程林.锂离子电池状态估计与剩余寿命预测方法综述[J].发电技术,2023,44(1):1-17. [ 5 ] SMITH J,ZHANG R,LEE T.Limitations of Electrochemical Impedance Spectroscopy(EIS) in Diagnosing Lithium-ion Battery Degradation Mechanisms[J].Journal of Power Sources,2021,500:123-130. [ 6 ] ZHANG L,WANG Q,LIU Y.Internal Resistance as a Health Indicator for Li-ion Batteries:Challenges and Misinterpretations[J].Energy Storage Materials,2020, 28:45-53. [ 7 ] CHEN Y,HUANG F,WU X.Microstructural Degradation Mechanisms in Lithium-ion Batteries:A Review[J]. Advanced Energy Materials,2022,12(15):2003456. [ 8 ] WANG H,ZHOU M,TANG S.Correlation Between Electrode Cracking and Capacity Fade in Commercial Li-ion Batteries[J].Nature Communications,2023,14: 789. [ 9 ] 柯胜蓝,吕亚帆,董玉军,等.一种电池健康状态评价方法及系统:2023112796209[P].2024-01-23. [10] 杨汉祥,郑小腾,陈央波,等.基于高分辨率工业 CT 的场(厂)内机动车用进水锂电池检测研究[J].CT 理论与应用研究(中英文),2024,33(2):159-166. [11] MüLLER S,SCHMIDT A,KRAUSE D.High-resolution Industrial CT for In-situ Analysis of Lithium-ion Battery Microstructures[J].ACS Applied Materials& Interfaces,2021,13(30):35672-35683. [12] FINEGAN D P,SCHEEL M,ROBINSON J B.Multiscale Imaging of Lithium-ion Battery Degradation[J].Joule, 2019,3(4):950-962. [13] LIU X,CHEN Z,LI J.Safety Risks in Second-life Lithium-ion Batteries:A Critical Review[J].Renewable and Sustainable Energy Reviews,2022,158:112035. [14] KIM S,PARK J,LEE H.Nondestructive Testing Strategies for Retired EV Batteries[J].IEEE Transactions on Transportation Electrification,2023,9(2):456-465. [15] 晋殿卫,顾则宇,张志宏.锂电池健康状态和剩余寿命预测算法研究[J].电力系统保护与控制,2023,51(1): 122-130. [16] 王斐.基于 LSTM 神经网络的锂电池 SOH 预测研究 [D].重庆:重庆邮电大学,2020. [17] 魏梓轩,韩晓娟,李炫.基于深度神经网络的梯次利用电池健康状态评估[J].太阳能学报,2022,43(5):518-524. [18] 赵显赫,耿光超,龚裕仲,等.基于深度学习的储能锂离子电池健康状态估计[J].能源工程,2022,42(4):28-35. [19] 李晓阳,谢恒义,韩贞辉,等.卷积神经网络在图像识别技术中的应用研究[J].能源与环保,2020,42(6):73-76. [20] OTSU N.A Threshold Selection Method from Graylevel Histograms[J].IEEE Transactions on Systems,Man, and Cybernetics,1979,9(1):62-66.

基于 CNN-LSTM 融合模型的动力电池健康状态精准评价研究

Research on Accurate Evaluation of Power Battery Health State Based on CNN-LSTM Hybrid Model

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 0

编辑推荐

Metrics

基于 CNN-LSTM 融合模型的动力电池 健康状态精准评价研究

Research on Accurate Evaluation of Power Battery Health State Based on CNN-LSTM Hybrid Model

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 0

编辑推荐

Metrics

基于 CNN-LSTM 融合模型的动力电池健康状态精准评价研究