基于自适应变分模态分解与重构的全钒液流电池健康状态预测

毛恒山; 刘浩骥; 刘晓杰; 王杰; 李伟; 胡炜昊; 韩文杰; 邱杰; 王枭; 熊斌宇

doi:10.19799/j.cnki.2095-4239.2025.0962

您当前的位置：

首页 >

文章列表页 >

基于自适应变分模态分解与重构的全钒液流电池健康状态预测

储能测试与评价 | 更新时间：2026-04-29

- 基于自适应变分模态分解与重构的全钒液流电池健康状态预测
- State of health prediction for vanadium flow batteries using adaptive variational mode decomposition and reconstruction
- 储能科学与技术 2026年15卷第4期页码：1412-1424
- 作者机构：
  
  1.中电建新能源集团股份有限公司，北京 100101
  2.武汉理工大学自动化学院，湖北武汉 430000
- 作者简介：
  
  毛恒山（1974—），男，硕士，工程师，研究方向为新型储能，E-mail：maohengshan@powerchina.cn；
  熊斌宇，教授，研究方向为VFB系统建模，E-mail：bxiong2@whut.edu.cn。
- 基金信息：
  
  2024年度中国电建新型储能研究中心定向委托计划(DJ-PTZX-2024-01);国家自然科学基金(52177221)
- DOI：10.19799/j.cnki.2095-4239.2025.0962
  中图分类号： TM 911
- 收稿：2025-10-27，
  
  修回：2025-11-22，
  
  纸质出版：2026-04-28
- 稿件说明：
移动端阅览
毛恒山, 刘浩骥, 刘晓杰, 等. 基于自适应变分模态分解与重构的全钒液流电池健康状态预测[J]. 储能科学与技术, 2026, 15(4): 1412-1424.

MAO Hengshan, LIU Haoji, LIU Xiaojie, et al. State of health prediction for vanadium flow batteries using adaptive variational mode decomposition and reconstruction[J]. Energy Storage Science and Technology, 2026, 15(4): 1412-1424.
毛恒山, 刘浩骥, 刘晓杰, 等. 基于自适应变分模态分解与重构的全钒液流电池健康状态预测[J]. 储能科学与技术, 2026, 15(4): 1412-1424. DOI： 10.19799/j.cnki.2095-4239.2025.0962.

MAO Hengshan, LIU Haoji, LIU Xiaojie, et al. State of health prediction for vanadium flow batteries using adaptive variational mode decomposition and reconstruction[J]. Energy Storage Science and Technology, 2026, 15(4): 1412-1424. DOI： 10.19799/j.cnki.2095-4239.2025.0962.

摘要

全钒液流电池(vanadium flow batteries，VFBs)在长期运行过程中，常因电解液体积失衡而引起异常容量衰减。因此对其健康状态(state of health，SOH)进行准确预测，对维持系统稳定运行至关重要。本研究基于电池循环老化数据，提出采用自适应变分模态分解与重构(adaptive variational mode decomposition and reconstruction，AVMDR)方法，以处理电池老化过程中的容量再生现象，并将其应用于SOH时间序列分析。基于相关性分析，重构了表征容量再生特征的波动函数

(

)与表征主体容量衰减趋势的主趋势函数

(

)。进而结合两者特性，分别采用长短期记忆网络(long short-term memory，LSTM)和Transformer模型，构建融合神经网络(integrated neural network，INN)模型。此外，通过概率分布计算以处理预测结果的不确定性问题。通过实验所得老化数据验证了所提融合模型的可行性与有效性。结果表明：该模型在多时间尺度下的预测均方根误差(RMSE)可保持在0.45%以内，在精度与稳定性方面均优于其他模型。

Abstract

Vanadium flow batteries (VFBs) are susceptible to abnormal capacity decay due to electrolyte volume imbalance during long-term operation. Therefore

accurate state of health (SOH) prediction is essential for maintaining system stability. In this study

a battery cycle aging data-based method based on adaptive variational mode decomposition and reconstruction (AVMDR) is proposed to address the capacity regeneration phenomenon observed during battery aging. The proposed method is applied to state-of-health time-series analysis. Correlation analysis is employed

to reconstruct a fluctuation function

(

) characterizing the capacity regeneration features and a main trend function

(

) representing the dominant capacity decay trend. An integrated neural network (INN) model is then constructed by employing a long short-term memory (LSTM) network and a Transformer model to handle the distinct characteristics of functions

(

) and

(

)

respectively. Furthermore

probability distribution calculations are performed to address the uncertainties in the prediction outcomes. The feasibility and effectiveness of the proposed hybrid model are validated using experimental aging data. Results demonstrate that the model maintains a root mean square error below 0.45% across multiple time scales

outperforming conventional models in both accuracy and stability.

关键词

Keywords

references

金阳. 新能源电力系统中的储能技术研究[J]. 低碳世界, 2023, 13(11): 49-51. DOI:10.3969/j.issn.2095-2066.2023.11.017.

JIN Y. Research on energy storage technology in new energy power system[J]. Low Carbon World, 2023, 13(11): 49-51. DOI:10.3969/j.issn.2095-2066.2023.11.017.

李建林, 梅岩竹, 王茜, 等. 全钒液流电池建模研究现状及展望[J]. 储能科学与技术, 2025, 14(12): 4618-4631.

LI J L, MEI Y Z, WANG Q, et al. Current status of vanadium redox flow battery modeling and research advances in data-driven approaches[J]. Energy Storage Science and Technology, 2025, 14(12): 4618-4631.

严腾飞, 曾义凯, 许刚. 全钒液流电池电解液制备技术研究进展[J]. 制冷与空调(四川), 2025, 39(3): 381-387, 438. DOI:10.3969/j.issn.1671-6612.2025.03.011.

YAN T F, ZENG Y K, XU G. Research progress on electrolyte preparation technologies for all-vanadium redox flow battery[J]. Refrigeration & Air Conditioning, 2025, 39(3): 381-387, 438. DOI:10.3969/j.issn.1671-6612.2025.03.011.

郑衍森, 苏黄盛, 梁业盈, 等. 风力发电用储能电池技术的发展现状及展望[J]. 电力电容器与无功补偿, 2025, 46(4): 1-12. DOI:10.14044/j.1674-1757.pcrpc.2025.04.001.

ZHENG Y S, SU H S, LIANG Y Y, et al. Development status and prospects of energy storage batteries for wind power generation[J]. Power Capacitor & Reactive Power Compensation, 2025, 46(4): 1-12. DOI:10.14044/j.1674-1757.pcrpc.2025.04.001.

王帅, 王远洋. 全钒氧化还原液流电池电解液的制备和性能优化研究进展[J]. 工业催化, 2025, 33(4): 26-33. DOI:10.3969/j.issn.1008-1143.2025.04.004.

WANG S, WANG Y Y. Research progress on preparation and performance optimization of all vanadium redox flow battery electrolyte[J]. Industrial Catalysis, 2025, 33(4): 26-33. DOI:10.3969/j.issn.1008-1143.2025.04.004.

岳孟, 郑琼, 阎景旺, 等. 液流电池流场结构设计与优化研究进展[J]. 化工进展, 2021, 40(9): 4853-4868. DOI:10.16085/j.issn.1000-6613.2021-0558.

YUE M, ZHENG Q, YAN J W, et al. Progress in flow field structure design and optimization for flow battery[J]. Chemical Industry and Engineering Progress, 2021, 40(9): 4853-4868. DOI:10.16085/j.issn.1000-6613.2021-0558.

熊瑞. 考虑水分子迁移的全钒液流电池容量失衡机理与建模方法研究[D]. 武汉: 武汉理工大学, 2022.XIONG R. Research on the vanadium redox flow battery capacity imbalance mechanism and its modeling considering water molecules migration[D]. Wuhan: Wuhan University of Technology, 2022.

SUN C X, CHEN J, ZHANG H M, et al. Investigations on transfer of water and vanadium ions across Nafion membrane in an operating vanadium redox flow battery[J]. Journal of Power Sources, 2010, 195(3): 890-897. DOI:10.1016/j.jpowsour.2009. 08.041.

杨皓凌, 徐昆誉, 张琪, 等. 钒液流电池中的改性Nafion膜[J]. 化学进展, 2023, 35(11): 1595-1612.

YANG H L, XU K Y, ZHANG Q, et al. Modified Nafion membrane in vanadium redox flow battery[J]. Progress in Chemistry, 2023, 35(11): 1595-1612.

SONG Y X, LI X R, XIONG J, et al. Electrolyte transfer mechanism and optimization strategy for vanadium flow batteries adopting a Nafion membrane[J]. Journal of Power Sources, 2020, 449: 227503. DOI:10.1016/j.jpowsour.2019.227503.

朱兆武, 张旭堃, 苏慧, 等. 全钒液流电池提高电解液浓度的研究与应用现状[J]. 储能科学与技术, 2022, 11(11): 3439-3446. DOI:10.19799/j.cnki.2095-4239.2022.0329.

ZHU Z W, ZHANG X K, SU H, et al. Research and application of increasing electrolyte concentration in all vanadium redox flow battery[J]. Energy Storage Science and Technology, 2022, 11(11): 3439-3446. DOI:10.19799/j.cnki.2095-4239.2022.0329.

李先锋, 张华民, 张洪章. 全钒液流电池用多孔离子传导隔膜的制备、结构调控及应用研究[C]//中国电子学会化学与物理电源技术分会, 中国化学与物理电源行业协会, 中国电工技术学会氢能发电装置专业委员会, 等. 第30届全国化学与物理电源学术年会论文集. 辽宁: 中科院大连化学物理研究所, 2013: 140.

XIONG B Y, ZHAO J Y, SU Y X, et al. State of charge estimation of vanadium redox flow battery based on sliding mode observer and dynamic model including capacity fading factor[J]. IEEE Transactions on Sustainable Energy, 2017, 8(4): 1658-1667. DOI:10.1109/TSTE.2017.2699288.

XIONG B Y, TANG J R, LI Y, et al. A flow-rate-aware data-driven model of vanadium redox flow battery based on gated recurrent unit neural network[J]. Journal of Energy Storage, 2023, 74: 109537. DOI:10.1016/j.est.2023.109537.

KHAKI B, DAS P. Sensorless parameter estimation of vanadium redox flow batteries in charging mode considering capacity fading[J]. Journal of Energy Storage, 2021, 33: 102033. DOI:10.1016/j.est.2020.102033.

杨斌. 全钒液流电池性能研究及SOC预测[D]. 西安: 西安理工大学, 2024.YANG B. Performance research and SOC prediction of all-vanadium redox flow battery[D]. Xi'an: Xi'an University of Technology, 2024.

曹鹏程, 吴和培, 刘浩, 等. 全钒液流电池电解液及隔膜研究进展[J]. 化工生产与技术, 2024, 30(6): 13-17, 37. DOI:10.3969/j.issn.1006-6829.2024.06.004.

CAO P C, WU H P, LIU H, et al. Research progress in all vanadium flow batteries[J]. Chemical Production and Technology, 2024, 30(6): 13-17, 37. DOI:10.3969/j.issn.1006-6829.2024.06.004.

李子鉴. 钒电池用双二氮杂萘酮聚芳醚砜离子交换膜[D]. 大连: 大连理工大学, 2024.LI Z J. Ion exchange membrane of bis-phthalazinone poly(aryl ether sulfone) for vanadium battery[D]. Dalian: Dalian University of Technology, 2024.

叶涛, 王怡君, 唐子龙, 等. 全钒液流电池电解液容量衰减及草酸恢复研究[J]. 储能科学与技术, 2025, 14(3): 1177-1186.

YE T, WANG Y J, TANG Z L, et al. Investigation of capacity fading in vanadium flow battery electrolytes and recovery via oxalic acid[J ] . Energy Storage Science and Technology, 2025, 14(3): 1177-1186.

罗碧灵. 基于优化VMD和IWOA的MMC子模块故障诊断方法研究[D]. 阜新: 辽宁工程技术大学, 2023.LUO B L. Research on fault diagnosis method for MMC submodules based on optimized VMD and IWOA[D]. Fuxin: Liaoning Technical University, 2023.

王丽玲, 孙晓波, 宋树平, 等. 基于QPSO-LSTM模型的锂电池剩余容量预测[J]. 机械与电子, 2024, 42(9): 52-56, 64. DOI:10.3969/j.issn.1001-2257.2024.09.008.

WANG L L, SUN X B, SONG S P, et al. Prediction of remaining capacity of lithium battery based on QPSO-LSTM model[J]. Machinery & Electronics, 2024, 42(9): 52-56, 64. DOI:10.3969/j.issn.1001-2257.2024.09.008.

朱铮, 戴辰, 蒋超, 等. 改进型CNN-LSTM深度学习神经网络的台区户变拓扑关系识别[J]. 电气自动化, 2024, 46(4): 93-95. DOI:10.3969/j.issn.1000-3886.2024.04.028.

ZHU Z, DAI C, JIANG C, et al. Recognition of topological relationship between substation transformers and users based on improved CNN-LSTM deep learning neural network[J]. Electrical Automation, 2024, 46(4): 93-95. DOI:10.3969/j.issn.1000-3886.2024.04.028.

WANG R H, LI C S, FU W L, et al. Deep learning method based on gated recurrent unit and variational mode decomposition for short-term wind power interval prediction[J]. IEEE Transactions on Neural Networks and Learning Systems, 2020, 31(10): 3814-3827. DOI:10.1109/TNNLS.2019.2946414.

陈逸嘉, 陶力, 李凯, 等. 基于IMIFS-VMD和ROA-LSTM的日前电价预测方法[J]. 自动化技术与应用, 2025, 44(9): 23-28. DOI:10.20033/j.1003-7241.(2025)09-0023-06.

CHEN Y J, TAO L, LI K, et al. Day-ahead electricity price forecasting method based on IMIFS-VMD and ROA-LSTM[J]. Techniques of Automation and Applications, 2025, 44(9): 23-28. DOI:10.20033/j.1003-7241.(2025)09-0023-06.

刘富强,刘为国,朱洪波,等. 基于KPCA与NRBO-Transformer的锂电池健康状态评估方法[J/OL].重庆理工大学学报(自然科学),1-12[2025-10-14].https://link.cnki.net/urlid/50.1205.T.20250904.1006.002.

池贝贝. 基于DEKF联合CNN-LSTM-Attention的锂电池SOC估计[J]. 软件, 2024, 45(7): 134-137. DOI:10.3969/j.issn.1003-6970. 2024.07.041.

CHI B B. SOC estimation of lithium batteries based on DEKF combined with CNN-LSTM-attention[J]. Computer Engineering & Software, 2024, 45(7): 134-137. DOI:10.3969/j.issn.1003-6970. 2024.07.041.

于小芳, 陈苏声, 周怡. 基于ELM锂离子电池RUL预测优化方法研究[J]. 环境技术, 2024, 42(6): 143-147. DOI:10.3969/j.issn.1004-7204.2024.06.032.

YU X F, CHEN S S, ZHOU Y. Research on RUL predictive optimization method based on ELM lithium-ion battery[J]. Environmental Technology, 2024, 42(6): 143-147. DOI:10.3969/j.issn.1004-7204.2024.06.032.

ZHANG Y W, TANG Q C, ZHANG Y, et al. Identifying degradation patterns of lithium ion batteries from impedance spectroscopy using machine learning[J]. Nature Communications, 2020, 11: 1706. DOI:10.1038/s41467-020-15235-7.

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

多健康指标的锂离子电池健康状态无分布区间估计

基于电化学阻抗谱几何解析的锂离子电池健康状态评估

基于白鹭群优化高斯过程回归的锂电池SOH估计方法

基于容量增量分析与VMD-GWO-KELM的锂电池健康状态估计

基于短时随机充电数据和优化卷积神经网络的锂电池健康状态估计