人工智能在飞轮储能中的应用

doi:10.19799/j.cnki.2095-4239.2025.0658

储能科学与技术 ›› 2025, Vol. 14 ›› Issue (8): 3019-3027.doi: 10.19799/j.cnki.2095-4239.2025.0658

• 短时高频高功率储能专辑 • 上一篇下一篇

人工智能在飞轮储能中的应用

魏路¹^,²(), 冷至益¹^,³^,⁴, 叶佳¹^,³^,⁴, 徐玉杰¹^,²^,³, 陈海生¹^,²^,³()

^1.中国科学院工程热物理研究所，北京 100190
^2.长时规模储能重点实验室(中国科学院)，北京 100190
^3.中国科学院大学，北京 100049
^4.中科南京未来能源系统研究院，江苏南京 211135

收稿日期:2025-07-22 修回日期:2025-08-01 出版日期:2025-08-28 发布日期:2025-08-18
通讯作者: 陈海生 E-mail:weilu@iet.cn;chen_hs@iet.cn
作者简介:魏路（1983—），男，硕士，工程师，主要研究方向为电力电子与电力传动，E-mail：weilu@iet.cn；
基金资助:
国家重点研发计划资助项目(2024YFB2408400)

Application of artificial intelligence in flywheel energy storage

Lu WEI¹^,²(), Zhiyi LENG¹^,³^,⁴, Jia YE¹^,³^,⁴, Yujie XU¹^,²^,³, Haisheng CHEN¹^,²^,³()

^1.Institute of Engineering Thermophysics, Chinese Academy of Science, Beijing 100190, China
^2.Key Laboratory of Long-Duration and Large-Scale Energy Storage (Chinese Academy of Sciences), Beijing 100190, China
^3.University of Chinese Academy of Sciences, Beijing 100049, China
^4.Nanijing Institute of Future Energy System, Institute of Engineering Thermophysics, Chinese Academy of Sciences, Nanjing 211135, Jiangsu, China

Received:2025-07-22 Revised:2025-08-01 Online:2025-08-28 Published:2025-08-18
Contact: Haisheng CHEN E-mail:weilu@iet.cn;chen_hs@iet.cn

摘要/Abstract

摘要：

飞轮储能系统(FESS)凭借高功率密度、长寿命、快速响应和环境友好等特性，在电网调频、惯量支撑、高频调峰等领域具有突出优势。飞轮储能系统面临着低成本高可靠设计、高速永磁电机和磁悬浮控制稳定性、在线故障预测以及多机并联阵列控制等问题。本文通过对近期相关文献的探讨，综述了人工智能技术在飞轮储能系统设计优化、电机控制、磁悬浮控制、并网控制及故障诊断等环节的应用，着重介绍了神经网络等算法在复合材料转子建模分析、永磁同步电机多参数协同优化设计、永磁同步电机多工况效率优化与转速观测、电磁轴承控制器算法、并网鲁棒性与分布式协同控制、调频控制策略、轴承故障诊断与预警等技术方向中的应用，并讨论大模型结合、多技术协同优化等未来发展方向，期望为飞轮储能系统的智能化研究和发展提供参考。

关键词: 飞轮储能, 人工智能, 调频, 故障诊断

Abstract:

Flywheel energy storage systems (FESSs) offer outstanding advantages in grid frequency regulation, inertia support, high-frequency peak shaving, and other applications due to their high power density, long service life, rapid responses, and environmental friendliness. However, FESSs face challenges related to achieving cost effectiveness and design reliability, stability of high-speed permanent magnet motors, control of magnetic suspension control, online prediction of faults, and control of multimachine parallel arrays. This paper reviews recent literature on the application of artificial intelligence (AI) technologies in key areas of FESSs, including design optimization, motor control, magnetic suspension control, grid-connected control, and fault diagnosis. Particular focus is given to the use of algorithms such as neural networks in several technical domains: modeling and analysis of composite material rotors, multiparameter collaborative optimization of permanent magnet synchronous motors (PMSMs), efficiency optimization and speed estimation of PMSMs under varying conditions, electromagnetic bearing control algorithms, grid-connected system robustness and distributed collaborative control, frequency regulation strategies, and bearing fault diagnosis and early warning. Future development directions such as the integration of large AI models and multitechnology collaborative optimization are also discussed. These insights aim to support intelligent research and development in FESSs.

Key words: flywheel energy storage, artificial intelligence, frequency regulation, fault diagnosis

中图分类号:

TK 02

魏路, 冷至益, 叶佳, 徐玉杰, 陈海生. 人工智能在飞轮储能中的应用[J]. 储能科学与技术, 2025, 14(8): 3019-3027.

Lu WEI, Zhiyi LENG, Jia YE, Yujie XU, Haisheng CHEN. Application of artificial intelligence in flywheel energy storage[J]. Energy Storage Science and Technology, 2025, 14(8): 3019-3027.

参考文献 47

[1]	OULD AMROUCHE S, REKIOUA D, REKIOUA T, et al. Overview of energy storage in renewable energy systems[J]. International Journal of Hydrogen Energy, 2016, 41(45): 20914-20927. DOI: 10.1016/j.ijhydene.2016.06.243.
[2]	AKINYELE D O, RAYUDU R K. Review of energy storage technologies for sustainable power networks[J]. Sustainable Energy Technologies and Assessments, 2014, 8: 74-91. DOI: 10.1016/j.seta.2014.07.004.
[3]	戴兴建, 魏鲲鹏, 张小章, 等. 飞轮储能技术研究五十年评述[J]. 储能科学与技术, 2018, 7(5): 765-782. DOI: 10.12028/j.issn.2095-4239.2018.0083.
	DAI X J, WEI K P, ZHANG X Z, et al. A review on flywheel energy storage technology in fifty years[J]. Energy Storage Science and Technology, 2018, 7(5): 765-782. DOI: 10.12028/j.issn.2095-4239.2018.0083.
[4]	BOLUND B, BERNHOFF H, LEIJON M. Flywheel energy and power storage systems[J]. Renewable and Sustainable Energy Reviews, 2007, 11(2): 235-258. DOI: 10.1016/j.rser.2005.01.004.
[5]	LI Y, LI Y B, JI P F, et al. Development of energy storage industry in China: A technical and economic point of review[J]. Renewable and Sustainable Energy Reviews, 2015, 49: 805-812. DOI: 10.1016/j.rser.2015.04.160.
[6]	PEARRE N S, SWAN L G. Technoeconomic feasibility of grid storage: Mapping electrical services and energy storage technologies[J]. Applied Energy, 2015, 137: 501-510. DOI: 10.1016/j.apenergy.2014.04.050.
[7]	AMIRYAR M E, PULLEN K R. A review of flywheel energy storage system technologies and their applications[J]. Applied Sciences, 2017, 7(3): 286. DOI: 10.3390/app7030286.
[8]	解学芳. 人工智能时代的文化创意产业智能化创新: 范式与边界[J]. 同济大学学报(社会科学版), 2019, 30(1): 42-51.
	XIE X F. Intelligent innovation of cultural and creative industries in the age of artificial intelligence: Paradigms and boundaries[J]. Journal of Tongji University (Social Science Section), 2019, 30(1): 42-51.
[9]	孙晔, 吴飞扬. 人工智能的研究现状及发展趋势[J]. 价值工程, 2013, 32(28): 5-7. DOI: 10.14018/j.cnki.cn13-1085/n.2013.28.137.
	SUN Y, WU F Y. Research status and development tendency of artificial intelligence[J]. Value Engineering, 2013, 32(28): 5-7. DOI: 10.14018/j.cnki.cn13-1085/n.2013.28.137.
[10]	王飞跃. 我国生成式人工智能的发展现状与趋势[J]. 人民论坛, 2025(2): 21-26.
	WANG F Y. Development status and trend of generative artificial intelligence in China[J]. People's Tribune, 2025(2): 21-26.
[11]	SINCHUK Y, SHISHKINA O, GUEGUEN M, et al. X-ray CT based multi-layer unit cell modeling of carbon fiber-reinforced textile composites: Segmentation, meshing and elastic property homogenization[J]. Composite Structures, 2022, 298: 116003. DOI: 10.1016/j.compstruct.2022.116003.
[12]	WANG Z F, WANG S, MA C W, et al. The prediction of homogenized effective properties of continuous fiber composites based on a deep transfer learning approach[J]. Composites Science and Technology, 2025, 262: 111050. DOI: 10.1016/j.compscitech.2025.111050.
[13]	HUANG S, SUN T, WANG S H, et al. The application of genetic algorithm in the structural optimization of permanent magnet synchronous motor[M]//The Proceedings of the 9th Frontier Academic Forum of Electrical Engineering. Singapore: Springer Singapore, 2021: 831-839. DOI: 10.1007/978-981-33-6609-1_76.
[14]	JIN Y X, WANG A Y, WANG T, et al. Optimal design of loss of permanent magnet synchronous motor based on particle swarm optimization[C]//2018 IEEE Student Conference on Electric Machines and Systems. December 14-16, 2018, Huzhou, China. IEEE, 2018: 1-4. DOI: 10.1109/SCEMS.2018.8624873.
[15]	SUN M X, XU Y L. Comprehensive optimization design of axial-flux permanent magnet synchronous machine for large-capacity flywheel energy storage system[C]//2024 IEEE 21st Biennial Conference on Electromagnetic Field Computation (CEFC). June 2-5, 2024, Jeju, Korea, Republic of. IEEE, 2024: 1-2. DOI: 10.1109/CEFC61729.2024.10585944.
[16]	朱迪, 赵杨阳, 艾邓鑫, 等. 基于遗传算法的飞轮储能电机多工况效率优化[J]. 储能科学与技术, 2024, 13(10): 3582-3592. DOI: 10.19799/j.cnki.2095-4239.2024.0249.
	ZHU D, ZHAO Y Y, AI D X, et al. Efficiency optimization of PMSM in flywheel energy storage under multiple working conditions based on genetic algorithm[J]. Energy Storage Science and Technology, 2024, 13(10): 3582-3592. DOI: 10.19799/j.cnki.2095-4239.2024.0249.
[17]	ZOLFAGHARI M, TAHER S A, MUNUZ D V. Neural network-based sensorless direct power control of permanent magnet synchronous motor[J]. Ain Shams Engineering Journal, 2016, 7(2): 729-740. DOI: 10.1016/j.asej.2016.01.002.
[18]	SHANTHI R, KALYANI S, DEVIE P M. RETRACTED ARTICLE: Design and performance analysis of adaptive neuro-fuzzy controller for speed control of permanent magnet synchronous motor drive[J]. Soft Computing, 2021, 25(2): 1519-1533. DOI: 10.1007/s00500-020-05236-5.
[19]	ZHANG H Y. Modeling permanent magnet synchronous motor system in electrical automation engineering based on adaptive neuro-fuzzy inference system[J]. Advanced Materials Research, 2013, 676: 297-301. DOI: 10.4028/www.scientific.net/amr.676.297.
[20]	邹晋彬, 易辉阳, 张鹏辉, 等. 基于Q-MPC的径向四自由度磁悬浮轴承控制策略[J]. 轴承, 2024(7): 36-44. DOI: 10.19533/j.issn1000-3762.2024.07.005.
	ZOU J B, YI H Y, ZHANG P H, et al. Control strategy of radial four degree of freedom magnetic bearings based on Q-MPC[J]. Bearing, 2024(7): 36-44. DOI: 10.19533/j.issn1000-3762.2024. 07.005.
[21]	KANG C F, LI K Y. Research on intelligent control system of permanent magnet motor for high-speed flywheel energy storage system based on deep learning[J]. Journal of Physics: Conference Series, 2024, 2729(1): 012001. DOI: 10.1088/1742-6596/2729/1/012001.
[22]	FITTRO R L, PANG D C, ANAND D K. Neural network controller development for a magnetically suspended flywheel energy storage system[C]//Proceedings of the NASA Langley Research Center, Second International Symposium on Magnetic Suspension Technology, Part 1, F, 1994.
[23]	ZHANG W Y, GUO F. Research on sensorless technology of a magnetic suspension flywheel battery based on a genetic BP neural network[J]. Actuators, 2025, 14(4): 174. DOI: 10.3390/act14040174.
[24]	ZHANG W Y, JI H T. High stability control of a magnetic suspension flywheel based on SA-BPNN and CNN+LSTM+ATTENTION[J]. Machines, 2024, 12(10): 710. DOI: 10.3390/machines12100710.
[25]	KANASE-PATIL A B, KALDATE A P, LOKHANDE S D, et al. A review of artificial intelligence-based optimization techniques for the sizing of integrated renewable energy systems in smart cities[J]. Environmental Technology Reviews, 2020, 9(1): 111-136. DOI: 10.1080/21622515.2020.1836035.
[26]	ZHAO D, SUN S Y, MOHAMMADZADEH A, et al. Adaptive intelligent model predictive control for microgrid load frequency[J]. Sustainability, 2022, 14(18): 11772. DOI: 10.3390/su141811772.
[27]	CHENG B, ZHANG W, YE M, et al. Flywheel energy storage control based on recurrent fuzzy neural network[C]//2010 8th World Congress on Intelligent Control and Automation. July 7-9, 2010, Jinan. IEEE, 2010: 4584-4589. DOI: 10.1109/WCICA. 2010.5554118.
[28]	王宁, 曲建真, 张志强, 等. 基于深度强化学习的轨交飞轮储能系统能量管理[J]. 科技创新与应用, 2025, 15(2): 30-33, 38. DOI: 10.19981/j.CN23-1581/G3.2025.02.006.
	WANG N, QU J Z, ZHANG Z Q, et al. Energy management of rail flywheel energy storage system based on deep reinforcement learning[J]. Technology Innovation and Application, 2025, 15(2): 30-33, 38. DOI: 10.19981/j.CN23-1581/G3.2025.02.006.
[29]	ZHOU J, JIA Y B, SUN C Y. Flywheel energy storage system controlled using tube-based deep Koopman model predictive control for wind power smoothing[J]. Applied Energy, 2025, 381: 125117. DOI: 10.1016/j.apenergy.2024.125117.
[30]	李佳玉. 基于多智能体协同的飞轮储能系统先进控制研究[D]. 北京: 华北电力大学, 2024. DOI: 10.27140/d.cnki.ghbbu.2024.000086.
	LI J Y. Research on advanced control of flywheel energy storage system based on multi-agent collaboration[D]. Beijing: North China Electric Power University, 2024. DOI: 10.27140/d.cnki.ghbbu.2024.000086.
[31]	季雯雯. 基于LSTM超短期风功率预测的风储联合系统一次调频研究[D]. 北京: 华北电力大学, 2024. DOI: 10.27140/d.cnki.ghbbu.2024.000572.
	JI W W. Research on frequency regulation of wind-storage joint system based on LSTM ultra-short-term wind power prediction[D]. Beijing: North China Electric Power University, 2024. DOI: 10.27140/d.cnki.ghbbu.2024.000572.
[32]	谢黎龙, 李勇汇, 范培潇, 等. 基于深度强化学习的孤立多微电网系统频率和电压综合控制[J]. 电力自动化设备, 2024, 44(6): 118-126. DOI: 10.16081/j.epae.202311013.
	XIE L L, LI Y H, FAN P X, et al. Deep reinforcement learning-based integrated frequency and voltage control for isolated multi-microgrid system[J]. Electric Power Automation Equipment, 2024, 44(6): 118-126. DOI: 10.16081/j.epae.202311013.
[33]	LEI M Z, MENG K, FENG H N, et al. Flywheel energy storage controlled by model predictive control to achieve smooth short-term high-frequency wind power[J]. Journal of Energy Storage, 2023, 63: 106949. DOI: 10.1016/j.est.2023.106949.
[34]	WANG J Y, LYU C H, BAI Y L, et al. Optimal scheduling strategy for hybrid energy storage systems of battery and flywheel combined multi-stress battery degradation model[J]. Journal of Energy Storage, 2024, 99: 113208. DOI: 10.1016/j.est.2024.113208.
[35]	LAI H T T, TRUNG H D, LAI K L, et al. Enhancing grid stability through predictive control and fuzzy neural networks in flywheel energy storage systems integration[J]. International Journal of Modern Physics B, 2025, 39(6): 2540020. DOI: 10.1142/s021797922540020x.
[36]	魏乐, 李承霖, 房方, 等. 小样本下基于改进麻雀算法优化卷积神经网络的飞轮储能系统损耗[J]. 电网技术, 2025, 49(1): 366-372. DOI: 10.13335/j.1000-3673.pst.2023.2001.
	WEI L, LI C L, FANG F, et al. Optimization of convolutional neural network based on improved sparrow algorithm for flywheel energy storage system loss in small sample[J]. Power System Technology, 2025, 49(1): 366-372. DOI: 10.13335/j.1000-3673.pst.2023.2001.
[37]	PATHAK P K, YADAV A K, PADMANABAN S, et al. Fractional cascade LFC for distributed energy sources via advanced optimization technique under high renewable shares[J]. IEEE Access, 2022, 10: 92828-92842.
[38]	SHUBHAM, ROY S P, MEHTA R K, et al. A novel application of jellyfish search optimisation tuned dual-stage (1+PI)TID controller for microgrid employing electric vehicle[J]. International Journal of Ambient Energy, 2022, 43(1): 8408-8427. DOI: 10.1080/01430750.2022.2097952.
[39]	BHAVANISANKAR C, SUDHA K R. An adaptive technique to control the load frequency of hybrid distributed generation systems[J]. Soft Computing, 2019, 23(23): 12385-12400. DOI: 10.1007/s00500-019-03779-w.
[40]	HUYNH V V, NAQVI S, NGUYEN B L, et al. Robust super-twisting algorithm-based single-phase sliding mode frequency controller in power systems integrating wind turbines and energy storage systems[J]. Scientific Reports, 2025, 15: 19740. DOI: 10.1038/s41598-025-01407-2.
[41]	ÇELIK E, ÖZTÜRK N, HOUSSEIN E H. Improved load frequency control of interconnected power systems using energy storage devices and a new cost function[J]. Neural Computing and Applications, 2023, 35(1): 681-697. DOI: 10.1007/s00521-022-07813-1.
[42]	魏乐, 季雯雯, 刘渝斌, 等. 基于长短期记忆神经网络超短期风功率预测的风储联合系统一次调频[J/OL]. 现代电力, 2025: 1-14. (2025-05-07). https://link.cnki.net/doi/10.19725/j.cnki.1007-2322.2024.0057.
	WEI L, JI W W, LIU Y B, et al. Primary frequency regulation strategy for a wind-storage-joint system based on LSTM ultra-short-term power generation prediction[J/OL]. Modern Electric Power, 2025: 1-14. (2025-05-07). https://link.cnki.net/doi/10.19725/j.cnki.1007-2322.2024.0057.
[43]	颜廷鑫. 飞轮储能径向轴承的状态检测与故障诊断研究[D]. 秦皇岛: 燕山大学, 2020. DOI: 10.27440/d.cnki.gysdu.2020.000745.
	YAN T X. Study on state detection and fault diagnosis of radial bearing for flying wheel energy storage[D]. Qinhuangdao: Yanshan University, 2020. DOI: 10.27440/d.cnki.gysdu.2020.000745.
[44]	吴朝辉. 飞轮储能系统故障诊断及智能预警方法研究[D]. 北京: 华北电力大学, 2023. DOI: 10.27140/d.cnki.ghbbu.2023.001509.
	WU C H. Research on fault diagnosis and intelligent early warning method of flywheel energy storage system[D]. Beijing: North China Electric Power University, 2023. DOI: 10.27140/d.cnki.ghbbu.2023.001509.
[45]	HE D Q, LIU C Y, JIN Z Z, et al. Fault diagnosis of flywheel bearing based on parameter optimization variational mode decomposition energy entropy and deep learning[J]. Energy, 2022, 239: 122108. DOI: 10.1016/j.energy.2021.122108.
[46]	NI Q, JI J C, FENG K, et al. A fault information-guided variational mode decomposition (FIVMD) method for rolling element bearings diagnosis[J]. Mechanical Systems and Signal Processing, 2022, 164: 108216. DOI: 10.1016/j.ymssp.2021.108216.
[47]	WANG H L, WU F, ZHANG L. Application of variational mode decomposition optimized with improved whale optimization algorithm in bearing failure diagnosis[J]. Alexandria Engineering Journal, 2021, 60(5): 4689-4699. DOI: 10.1016/j.aej.2021.03.034.

人工智能在飞轮储能中的应用

Application of artificial intelligence in flywheel energy storage

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献 47

相关文章 15

编辑推荐

Metrics

本文评价

[1]	马骞, 肖亮, 程冰, 高琴, 刘春晓, 朱益华, 李成翔. 基于强化学习-模型预测控制（RL-MPC）的分布式储能协同一次调频控制方法[J]. 储能科学与技术, 2025, 14(8): 3138-3148.
[2]	徐鹤勇, 郑铁军, 丁圣权, 蒙飞, 张越, 杨家麒. 基于量子增强混合时空图神经网络的混合储能系统自适应频率调节方法[J]. 储能科学与技术, 2025, 14(8): 3149-3159.
[3]	李博文, 文贤馗, 范强, 古庭赟, 史正军, 张晓寅. MW级飞轮电机转子中空轴内通流散热实验研究[J]. 储能科学与技术, 2025, 14(8): 2925-2931.
[4]	文贤馗, 李博文, 史正军, 叶华洋, 庞玲蓉, 张晓寅. 飞轮储能系统磁轴承电磁特性与温升特性分析[J]. 储能科学与技术, 2025, 14(8): 2932-2941.
[5]	孙召琴, 黎可, 杜杲娴, 胡晨, 牛萌, 朱真. 多端口能量路由器快速频率响应及稳定控制方法[J]. 储能科学与技术, 2025, 14(8): 2970-2982.
[6]	徐彩莹, 唐毓振, 李秋雨, 杨浩岳, 陈洋, 杨恒昭. 用于电力系统调频的超级电容储能系统[J]. 储能科学与技术, 2025, 14(8): 3078-3089.
[7]	魏江哲, 舒茂龙, 陈彦桥, 赵璐璐, 苏新凯, 刘牧阳. 考虑边际替代效应的火-储联合系统调频备用配置方法[J]. 储能科学与技术, 2025, 14(5): 2117-2129.
[8]	萨仁高娃, 邬超慧, 倪泽龙, 张悦, 姜新建, 田建宇. 基于强化学习的变参数PID的惯量飞轮有功控制策略[J]. 储能科学与技术, 2025, 14(5): 1982-1990.
[9]	王艺斐, 徐帆, 王亮, 戴兴建, 徐玉杰, 陈海生. 飞轮储能系统电机定子散热设计研究[J]. 储能科学与技术, 2025, 14(5): 1946-1953.
[10]	许庆祥, 滕伟, 秦润, 宋顺一, 柳亦兵, 梁双印. 电网调频飞轮储能系统并网能量管理与控制策略[J]. 储能科学与技术, 2025, 14(5): 2013-2022.
[11]	王滔, 毛田, 周保荣, 赵文猛, 华浩. 基于储能荷电状态的虚拟同步机控制探讨[J]. 储能科学与技术, 2025, 14(5): 2032-2034.
[12]	韩松. 基于大数据和人工智能的储能系统故障预测与诊断方法研究[J]. 储能科学与技术, 2025, 14(5): 2114-2116.
[13]	刘晓悦, 陈炎, 孙小菲. 飞轮阵列参与电网一次调频双层控制策略[J]. 储能科学与技术, 2025, 14(4): 1536-1547.
[14]	李石明. 人工智能在储能系统故障检测中的应用[J]. 储能科学与技术, 2025, 14(4): 1698-1700.
[15]	董文琦, 张东晖, 曹一凡, 宁照轩, 姜新建, 李明, 史学伟. 新型惯量飞轮与高速飞轮参与电网惯性响应与一次调频的控制策略[J]. 储能科学与技术, 2025, 14(3): 1224-1233.