基于分数阶阻抗模型的锂离子电池漏液故障诊断

刘伟斌; 黄志成; 易斌; 黄铭锐; 张梦麟; 郭豪腾

doi:10.19799/j.cnki.2095-4239.2026.0594

您当前的位置：

首页 >

文章列表页 >

基于分数阶阻抗模型的锂离子电池漏液故障诊断

储能测试与评价 | 更新时间：2026-10-22

- 基于分数阶阻抗模型的锂离子电池漏液故障诊断
- Diagnosis of lithium-ion battery leakage fault based on fractional-order model parameters
- 储能科学与技术 2026年15卷第10期页码：4148-4160
- 作者机构：
  
  1.广东电网有限责任公司东莞供电局，广东东莞 523000
  2.广东新型储能国家研究院有限公司，广东广州 510000
- 作者简介：
  
  刘伟斌（1986—），男，硕士，高级工程师，从事配电设备运行，E-mail：706360854@qq.com；
  郭豪腾，硕士，从事锂离子电池故障诊断技术与早期预警，E-mail：23126280@bjtu.edu.cn。
- 基金信息：
  
  南方电网科技项目［031900KC24040028(GDKJXM20240438)
- DOI：10.19799/j.cnki.2095-4239.2026.0594
  中图分类号： TM 911
- 收稿：2026-07-08，
  
  修回：2026-07-23，
  
  纸质出版：2026-10-28
- 稿件说明：
移动端阅览
刘伟斌, 黄志成, 易斌, 等. 基于分数阶阻抗模型的锂离子电池漏液故障诊断[J]. 储能科学与技术, 2026, 15(10): 4148-4160.

LIU Weibin, HUANG Zhicheng, YI bin, et al. Diagnosis of lithium-ion battery leakage fault based on fractional-order model parameters[J]. Energy Storage Science and Technology, 2026, 15(10): 4148-4160.
刘伟斌, 黄志成, 易斌, 等. 基于分数阶阻抗模型的锂离子电池漏液故障诊断[J]. 储能科学与技术, 2026, 15(10): 4148-4160. DOI： 10.19799/j.cnki.2095-4239.2026.0594.

LIU Weibin, HUANG Zhicheng, YI bin, et al. Diagnosis of lithium-ion battery leakage fault based on fractional-order model parameters[J]. Energy Storage Science and Technology, 2026, 15(10): 4148-4160. DOI： 10.19799/j.cnki.2095-4239.2026.0594.

摘要

随着电动汽车市场占有率的攀升，锂离子电池呈现出爆发式增长和规模化应用，安全仍然是电池面临的最大挑战。受制造工艺缺陷影响，商用锂离子圆柱电池顶盖处易在摩擦与挤压作用下破损，导致内部电解液接触空气并发生泄漏。锂离子电池电解液泄漏故障不仅影响单体能量的存储和释放，也可能会引起外短路，甚至导致起火的风险，对电池运行安全构成严重威胁。为准确识别漏液电池，本研究设计了电池电解液泄漏故障模拟实验，开展漏液电池和正常电池的循环老化和电化学阻抗谱测试，并基于交流阻抗谱数据建立了电池二阶分数阶等效阻抗模型。利用自适应遗传算法辨识了模型参数，揭示了正常电池与漏液电池阻抗模型参数的演变规律。最终，利用决策树分类算法对提取的故障特征样本进行训练与测试。为降低单次样本划分带来的随机性影响，本工作进行了100次重复样本划分与故障诊断，并计算准确率、精确率、召回率和特异性的平均值以评估模型分类性能。同时，设置不同的训练集与测试集分配比例，对各条件下的分类结果进行了对比分析。并且，将单一电池独立作为测试对象，其余电池作为训练样本开展跨电池故障诊断，以进一步验证所构建模型对未知电池的泛化能力，为锂离子电池安全预警奠定了重要基础。

Abstract

With the continuously increasing market share of electric vehicles

lithium-ion batteries are experiencing explosive growth and large-scale application. However

safety remains the greatest challenge in battery technology. Due to manufacturing defects

commercial cylindrical lithium-ion batteries are prone to damage of the top cover under friction or compression. Such damage leads to exposure of the internal electrolyte to air and its subsequent leakage. Electrolyte leakage not only affects the storage and release of energy in individual cells but also causes external short circuits and fire hazards

posing a serious threat to battery operational safety. Hence

it is critical to accurately identify leaking batteries. To this end

this study designed a simulation experiment to identify electrolyte leakage faults

conducted cycle aging tests and electrochemical impedance spectroscopy measurements on both leaking and normal batteries

and established a second-order fractional-order equivalent impedance model based on AC impedance data. Model parameters were identified using an adaptive genetic algorithm to reveal the evolution patterns of impedance parameters between normal and leaking batteries. Finally

a decision-tree classification algorithm was used to train and test the extracted fault feature samples. Further

100 repeated sample divisions and fault diagnoses were performed to reduce the randomness caused by single sample partitioning. The average values of accuracy

precision

recall

and specificity were calculated to evaluate the model's classification performance. Additionally

different training-to-testing set ratios were tested

and classification results obtained under various conditions were compared and analyzed. Furthermore

a cross-battery fault diagnosis was conducted by taking one battery as the testing object and the rest as training samples. The diagnosis results further validated the model's generalization capability for unknown batteries

laying a solid foundation for early warning systems for lithium-ion battery safety.

关键词

Keywords

references

HAN X B, LU L G, ZHENG Y J, et al. A review on the key issues of the lithium ion battery degradation among the whole life cycle[J]. eTransportation, 2019, 1: 100005. DOI:10.1016/j.etran. 2019.100005.

LI X Y, YUAN C G, WANG Z P, et al. Lithium battery state-of-health estimation and remaining useful lifetime prediction based on non-parametric aging model and particle filter algorithm[J]. eTransportation, 2022, 11: 100156. DOI:10.1016/j.etran. 2022.100156.

ZHANG Z P, LI Z, YANG X L, et al. Thermal management of parallel cells—Methodology based on non-linear dynamics[J]. eTransportation, 2022, 13: 100187. DOI:10.1016/j.etran. 2022.100187.

孙敬延. "双碳"目标驱动下的能源结构优化与转型路径[J]. 生态与资源, 2024(11): 41-43.

SUN J Y. Energy structure optimization and transformation path driven by the "dual carbon" goal[J]. Ecology and Resources, 2024(11): 41-43.

WANG Y Y, LAI X, CHEN Q W, et al. Progress and challenges in ultrasonic technology for state estimation and defect detection of lithium-ion batteries[J]. Energy Storage Materials, 2024, 69: 103430. DOI:10.1016/j.ensm.2024.103430.

赖铱麟, 杨凯, 刘皓, 等. 锂离子电池安全预警方法综述[J]. 储能科学与技术, 2020, 9(6): 1926-1932.

LAI Y L, YANG K, LIU H, et al. Lithium-ion battery safety warning methods review[J]. Energy Storage Science and Technology, 2020, 9(6): 1926-1932.

DU X W, YANG B, LU Y, et al. Detection of electrolyte leakage from lithium-ion batteries using a miniaturized sensor based on functionalized double-walled carbon nanotubes[J]. Journal of Materials Chemistry C, 2021, 9(21): 6760-6765. DOI:10.1039/d1tc01069g.

WAN J, LIU C G, WANG X X, et al. Conductometric sensor for ppb-level lithium-ion battery electrolyte leakage based on Co/Pd-doped SnO 2 [J ] . Sensors and Actuators B: Chemical, 2023, 393: 134326. DOI:10.1016/j.snb.2023.134326.

吴磊, 吕桃林, 陈启忠, 等. 电化学阻抗谱测量与应用研究综述[J]. 电源技术, 2021, 45(9): 1227-1230. DOI:10.3969/j.issn.1002-087X. 2021.09.035.

WU L, LV T L, CHEN Q Z, et al. Review of measurement and application of electrochemical impedance spectroscopy[J]. Chinese Journal of Power Sources, 2021, 45(9): 1227-1230. DOI:10.3969/j.issn.1002-087X.2021.09.035.

冷晓伟, 戴作强, 郑莉莉, 等. 锂离子电池电化学阻抗谱研究综述[J]. 电源技术, 2018, 42(11): 1749-1752. DOI:10.3969/j.issn.1002-087X.2018.11.048.

LENG X W, DAI Z Q, ZHENG L L, et al. Review on electrochemical impedance spectroscopy of lithium-ion batteries[J]. Chinese Journal of Power Sources, 2018, 42(11): 1749-1752. DOI:10.3969/j.issn.1002-087X.2018.11.048.

口尧, 李日康, 王学远, 等. 基于电化学阻抗谱的锂离子电池等效电路模型参数辨识方法[J]. 机械与电子, 2021, 39(4): 33-38. DOI:10.3969/j.issn.1001-2257.2021.04.007.

KOU Y, LI R K, WANG X Y, et al. A parameter identification method of lithium-ion battery equivalent circuit model based on electrochemical impedance spectroscopy[J]. Machinery & Electronics, 2021, 39(4): 33-38. DOI:10.3969/j.issn.1001-2257. 2021.04.007.

申明尧, 韩萌, 杜诗语, 等. 数据流决策树集成分类算法综述[J]. 计算机应用与软件, 2022, 39(9): 1-10. DOI:10.3969/j.issn.1000-386x. 2022.09.001.

SHEN M Y, HAN M, DU S Y, et al. A survey of decision tree ensemble classification algorithms of data streams[J]. Computer Applications and Software, 2022, 39(9): 1-10. DOI:10.3969/j.issn.1000-386x.2022.09.001.

ZOU C F, ZHANG L, HU X S, et al. A review of fractional-order techniques applied to lithium-ion batteries, lead-acid batteries, and supercapacitors[J]. Journal of Power Sources, 2018, 390: 286-296. DOI:10.1016/j.jpowsour.2018.04.033.

贾先屹, 王顺利, 曹文, 等. 基于二阶RC等效电路和SR-DUKF算法的锂电池SOC估算研究[J]. 电源学报, 2024, 22(4): 236-242. DOI:10.13234/j.issn.2095-2805.2024.4.236.

JIA X Y, WANG S L, CAO W, et al. Research on estimation of lithium battery SOC based on second-order RC equivalent circuit and SR-DUKF algorithm[J]. Journal of Power Supply, 2024, 22(4): 236-242. DOI:10.13234/j.issn.2095-2805.2024.4.236.

吴健, 尹泽, 李豪, 等. 基于分数阶理论的锂离子电池高频等效电路模型[J]. 电工技术学报, 2021, 36(18): 3902-3910.

WU J, YIN Z, LI H, et al. High-frequency equivalent circuit model of lithium-ion battery based on fractional order theory[J]. Transactions of China Electrotechnical Society, 2021, 36(18): 3902-3910.

许志华. 锂电池的分数阶建模及状态估计方法研究[J]. 电源学报, 2025, 23(2): 266-273. DOI:10.13234/j.issn.2095-2805.2025.2.266.

XU Z H. Fractional-order modeling and state estimation method for lithium battery[J]. Journal of Power Supply, 2025, 23(2): 266-273. DOI:10.13234/j.issn.2095-2805.2025.2.266.

李明伟, 张伟峰, 王翼. 基于自适应遗传算法和SVM的电机故障辨识[J]. 包装工程, 2023, 44(S2): 92-96.

LI M W, ZHANG W F, WANG Y. Motor fault identification based on adaptive genetic algorithm and SVM[J]. Packaging Engineering, 2023, 44(S2): 92-96.

WANG Y J, LI M C, CHEN Z H. Experimental study of fractional-order models for lithium-ion battery and ultra-capacitor: Modeling, system identification, and validation[J]. Applied Energy, 2020, 278: 115736. DOI:10.1016/j.apenergy.2020.115736.

SAHU A R, PALEI S K, MISHRA A. Data-driven fault diagnosis approaches for industrial equipment: A review[J]. Expert Systems, 2024, 41(2): e13360. DOI:10.1111/exsy.13360.

WEN L, LI X Y, GAO L, et al. A new convolutional neural network-based data-driven fault diagnosis method[J]. IEEE Transactions on Industrial Electronics, 2018, 65(7): 5990-5998. DOI:10.1109/TIE.2017.2774777.

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

基于EIS的燃料电池故障在线诊断研究进展

三元锂电池低温充电老化特征及基于弛豫阻抗的析锂检测方法

基于双层机器学习的非对称超级电容器跨表征EIS参数估计

基于电化学阻抗谱和弛豫时间分析的铅酸电池老化状态诊断研究

基于改进TD3算法的储能系统多目标优化调度策略研究