抵御虚假数据注入攻击的网络化微电网多层级韧性控制框架

黄煜; 曹正阳; 胡松林; 岳东; 陈永华; 颜云松

doi:10.11999/JEIT250850

抵御虚假数据注入攻击的网络化微电网多层级韧性控制框架

doi: 10.11999/JEIT250850 cstr: 32379.14.JEIT250850

1.
南京邮电大学碳中和先进技术研究院南京 210023
2.
国电南瑞科技股份有限公司南京 211106

基金项目: 江苏省自然科学基金资助项目(BK20232026)

详细信息

作者简介:
黄煜：男，副教授，研究方向为信息物理能源电力系统网络化控制与优化

曹正阳：男，硕士研究生，研究方向为信息物理能源电力系统网络化控制与优化

胡松林：男，教授，研究方向为信息物理能源电力系统网络化控制与优化

岳东：男，教授，研究方向为信息物理能源电力系统网络化控制与优化

陈永华：男，高级工程师，研究方向为电力系统安全稳定控制

颜云松：男，高级工程师，研究方向为电力系统安全稳定控制

通讯作者:
胡松林　slhu621@njupt.edu.com

中图分类号: TM721; TP393.08; TN915.08
计量
- 文章访问数: 15
- HTML全文浏览量: 5
- PDF下载量: 0
- 被引次数: 0
出版历程
- 修回日期: 2026-04-08
- 录用日期: 2026-04-08
- 网络出版日期: 2026-04-25

A Multi-layer Resilient Control Framework for Networked Microgrids against False Data Injection Attacks

1.
Institute of Advanced Technology for Carbon Neutrality, Nanjing University of Posts and Telecommunications, Nanjing 210023, China
2.
NARI Technology Co., Ltd, Nanjing 211106, China

Funds: This work is supported by Natural Science Foundation of Jiangsu Province (No. BK20232026)

摘要

摘要: 针对网络化微电网在高渗透可再生能源和开放通信环境下易受虚假数据注入攻击(FDIA)影响、导致频率失稳与通信拥塞的问题，本文提出一种“事件触发通信—攻击观测—深度强化学习补偿”的多层级一体化韧性控制方法。该方法设计了基于频率—积分误差的自适应事件触发机制以削减冗余数据传输，并构建径向基函数未知输入观测器(RBF-UIO)实现对FDIA的高精度检测与状态估计。在此基础上，引入双重回放Q学习(DRQL)开展在线补偿优化，实现攻击抑制与性能恢复。进一步，在典型4节点分布式微电网平台上开展仿真验证，结果表明：所提方法在显著降低通信事件数的同时，有效提升了频率恢复速度与稳态精度，保证了虚假数据攻击下的安全稳定运行。
- 网络化微电网 /
- 径向基函数观测器 /
- 强化学习 /
- 虚假数据注入攻击 /
- 韧性控制
Abstract: Objective With the increasing penetration of renewable distributed energy and the growing reliance on cyber-physical infrastructures, networked microgrids (NMGs) have become highly vulnerable to false data injection attacks (FDIAs), which threaten frequency stability and system security. Traditional secondary control methods, constrained by limited communication resources and rigid sampling mechanisms, struggle to ensure resilient operation when facing stealthy cyber-attacks and dynamic disturbances. To address these challenges, it is imperative to design control frameworks that can jointly optimize communication efficiency, enhance attack detection, and guarantee rapid stability recovery. This study therefore develops a multi-layer resilient control strategy that integrates event-triggered communication, data-driven observation, and deep reinforcement learning, aiming to provide an effective solution for securing the stability of NMGs against sophisticated cyber threats. Methods The proposed ETC–RBF–DRQL framework integrates an event-triggered communication mechanism, a radial basis function (RBF) observer, and a deep reinforcement learning (DRQL) compensator to achieve resilient frequency control in networked microgrids under false data injection attacks (FDIAs). The event-triggered scheme reduces redundant data transmission while maintaining stability, and the RBF observer estimates system states and detects anomalous deviations. Upon detection, the DRQL module adaptively generates compensation signals to suppress attack impact and restore system stability. The framework is mathematically formulated within a modularized dynamic model of networked microgrids, ensuring provable stability under communication and attack constraints. Simulation experiments are conducted on a 4-node distributed microgrid testbed in MATLAB/Simulink, including diverse renewable energy sources and realistic communication links, to validate the effectiveness and scalability of the proposed approach. Results and Discussions The proposed ETC-RBF-DRQL framework was validated on a 4-node networked microgrid under FDIA scenarios. Simulation results show that the method achieves superior overall performance in frequency regulation, communication efficiency, and attack resilience. Specifically, the frequency deviation peak is reduced from 0.0218 Hz under periodic PI control to 0.0121 Hz, while the steady-state average deviation and fluctuation standard deviation are reduced to 0.0097 Hz and 0.0074 Hz, respectively (Fig.4, Table 2). Meanwhile, the average communication event rate decreases to 11.9 pkt·s^-1, corresponding to a 76.2% reduction compared with periodic sampling (Table 2). In addition, the proposed framework maintains reliable attack detection performance, with a detection rate of 91.5%, a false alarm rate of 4.8%, and an AUC of 0.968 (Table 2). These results demonstrate that the proposed method can effectively coordinate frequency recovery, communication saving, and FDIA mitigation in networked microgrids. Conclusions This paper investigates a multi-layer resilient control framework for networked microgrids under FDIAs and communication constraints. The proposed ETC–RBF–DRQL method integrates event-triggered communication, RBF-based attack detection, and dual Q-learning-based adaptive compensation, thereby achieving closed-loop coordination of anomaly detection, attack suppression, and frequency stability recovery. Simulation results on a 4-node networked microgrid demonstrate that, compared with traditional PI-based schemes, the proposed approach significantly reduces frequency deviation peaks and shortens recovery time, while effectively lowering communication overhead. Theoretical analysis further confirms its feasibility and stability under bounded estimation errors. Nevertheless, this study focuses on sensor-side FDIAs and simplified communication conditions; future work will extend to more complex multi-type attacks and hardware-in-the-loop validation to advance engineering applications.
- Networked Microgrids /
- Radial Basis Function Observer /
- Reinforcement Learning /
- False Data Injection Attack (FDIA) /
- Resilient Control

HTML全文

图 1 网络化微电网第i区频率控制闭环示意图

下载: 全尺寸图片幻灯片

图 2 ETC-RBF-DRQL一体化控制架构

下载: 全尺寸图片幻灯片

图 3 4节点网络化微电网系统拓扑结构

下载: 全尺寸图片幻灯片

图 4 节点2在FDIA攻击下的频率偏差响应曲线

下载: 全尺寸图片幻灯片

图 5 基于优先经验回放的双重Q学习收敛性曲线

下载: 全尺寸图片幻灯片

图 6 不同事件触发参数对频率恢复时间的敏感性分析

下载: 全尺寸图片幻灯片

表 1 4节点NMG系统关键物理与控制参数

参数	含义	数值/单位
$ {H}_{i} $	节点惯性常数(同步机/虚拟同步机)	1.5 s(MG1)，2.0 s(MG3)，0.8 s(MG2、MG4)
$ {D}_{i} $	负荷阻尼系数	0.8(各节点)
$ {R}_{i} $	下垂系数	0.04 Hz/p.u.(各节点)
$ {T}_{g,i} $	调速器时间常数	0.15 s(柴油机)，0.08 s(VSG 节点)
$ {T}_{c} $	通信采样周期/触发网格	20 ms
$ {\sigma }_{f}/{\sigma }_{z} $	事件触发阈值权重(频差/积分差)	1.2 / 1.2
$ \theta $	残差检测死区阈值	5×10^–4 p.u.
$ \overline{\epsilon } $	RBF近似误差上界	1×10^–3 p.u.

下载: 导出CSV

表 2 不同控制策略下的频率性能、通信负载与检测性能对比

策略	峰值超调量$ \|\Delta f{\|}_{\text{max}} $(Hz)	稳态平均偏差$ \|\overline{\Delta f}\| $(Hz)	稳态波动标准差$ \sigma $(Hz)	平均通信事件速率(pkt·s^–1)	通信节省率(%)	FDIA检测率(%)	误报率(%)	AUC
周期采样PI(基准)	0.0218	0.0153	0.0102	50.0	0.0	—	—	—
ETC+PI	0.0165	0.0124	0.0091	16.8	66.4	—	—	—
ETC+RBF+PI	0.0148	0.0109	0.0083	14.3	71.4	91.5	4.8	0.968
所提ETC-RBF-DRQL	0.0121	0.0097	0.0074	11.9	76.2	91.5	4.8	0.968

下载: 导出CSV

参考文献(29)

[1]	刘鑫蕊, 张明超, 王睿, 等. 基于一致性算法的城市微网群分层协同控制[J]. 电力系统自动化, 2022, 46(17): 65–73. doi: 10.7500/AEPS20220228007. LIU Xinrui, ZHANG Mingchao, WANG Rui, et al. Hierarchical collaborative control of urban microgrid cluster based on consensus algorithm[J]. Automation of Electric Power Systems, 2022, 46(17): 65–73. doi: 10.7500/AEPS20220228007.
[2]	孟潇潇, 邵冰冰, 韩平平, 等. 基于背靠背变流器柔性互联的微网群分层协同恢复控制策略[J]. 中国电机工程学报, 2023, 43(20): 7812–7826. doi: 10.13334/j.0258-8013.pcsee.221457. MENG Xiaoxiao, SHAO Bingbing, HAN Pingping, et al. Hierarchical cooperative recovery control strategy for flexible interconnected microgrid cluster based on back-to-back converters[J]. Proceedings of the CSEE, 2023, 43(20): 7812–7826. doi: 10.13334/j.0258-8013.pcsee.221457.
[3]	陈郁林, 谷雨润, 闫云凤, 等. 有向通信拓扑下基于分布式触发控制的微电网二次控制方法[J]. 电子与信息学报, 2022, 44(11): 3806–3814. doi: 10.11999/JEIT220866. CHEN Yulin, GU Yurun, YAN Yunfeng, et al. Secondary control methods based on distributed event-triggered control in microgrids under directed communication network[J]. Journal of Electronics & Information Technology, 2022, 44(11): 3806–3814. doi: 10.11999/JEIT220866.
[4]	王本斐, 张荣辉, 冯国栋, 等. 基于事件触发的直流微电网无差拍预测控制[J]. 自动化学报, 2024, 50(3): 475–485. doi: 10.16383/j.aas.c210585. WANG Benfei, ZHANG Ronghui, FENG Guodong, et al. Event-triggered deadbeat predictive control for DC microgrid[J]. Acta Automatica Sinica, 2024, 50(3): 475–485. doi: 10.16383/j.aas.c210585.
[5]	JAFARI M, RAHMAN M A, and PAUDYAL S. Optimal false data injection attack against load-frequency control in power systems[J]. IEEE Transactions on Information Forensics and Security, 2023, 18: 5200–5212. doi: 10.1109/TIFS.2023.3305868.
[6]	BAHARIZADEH M, GOLSORKHI M S, and SAVAGHEBI M. Secondary control with reduced communication requirements for accurate reactive power sharing in AC microgrids[J]. IET Smart Grid, 2023, 6(6): 638–652. doi: 10.1049/stg2.12127.
[7]	CHENG Zihao, BU Xuhui, LIANG Jiaqi, et al. Indirect–direct secure load frequency control against false data injection attacks[J]. IEEE Transactions on Industrial Informatics, 2024, 20(3): 4850–4862. doi: 10.1109/TII.2023.3330235.
[8]	DANG Lujuan, CHEN Badong, WANG Shiyuan, et al. Robust power system state estimation with minimum error entropy unscented Kalman filter[J]. IEEE Transactions on Instrumentation and Measurement, 2020, 69(11): 8797–8808. doi: 10.1109/TIM.2020.2999757.
[9]	李建华. 能源关键基础设施网络安全威胁与防御技术综述[J]. 电子与信息学报, 2020, 42(9): 2065–2081. doi: 10.11999/JEIT191055. LI Jianhua. Overview of cyber security threats and defense technologies for energy critical infrastructure[J]. Journal of Electronics & Information Technology, 2020, 42(9): 2065–2081. doi: 10.11999/JEIT191055.
[10]	KHALILI J, DEHKORDI N M, and HAMZEH M. Distributed event-triggered secondary frequency control of islanded AC microgrids under cyber attacks with input time delay[J]. International Journal of Electrical Power & Energy Systems, 2022, 143: 108506. doi: 10.1016/j.ijepes.2022.108506.
[11]	HOSSAIN M, PENG Chen, SUN Hongtao, et al. Bandwidth allocation-based distributed event-triggered LFC for smart grids under hybrid attacks[J]. IEEE Transactions on Smart Grid, 2022, 13(1): 820–830. doi: 10.1109/TSG.2021.3118801.
[12]	ZHAO Guanglei, JIN Lanqing, CUI Hailong, et al. Distributed adaptive dynamic event-triggered secondary control for islanded microgrids with disturbances[J]. IEEE Transactions on Smart Grid, 2023, 14(6): 4268–4281. doi: 10.1109/TSG.2023.3264979.
[13]	MOHAMMADI K, AZIZI E, CHOI J, et al. Asynchronous periodic distributed event-triggered voltage and frequency control of microgrids[J]. IEEE Transactions on Power Systems, 2021, 36(5): 4524–4538. doi: 10.1109/TPWRS.2021.3059158.
[14]	ZHANG Xianming, HAN Qinglong, GE Xiaohua, et al. An overview of recent advances in event-triggered control[J]. Science China Information Sciences, 2025, 68(6): 161201. doi: 10.1007/s11432-024-4437-9.
[15]	SHAFEI H, FARHANGI M, LI Li, et al. A novel cyber-attack detection and mitigation for coupled power and information networks in microgrids using distributed sliding mode unknown input observer[J]. IEEE Transactions on Smart Grid, 2025, 16(2): 1667–1681. doi: 10.1109/TSG.2024.3475468.
[16]	ZHUANG Xincheng, TIAN Yang, WANG Haoping, et al. Neural network adaptive observer design for nonlinear systems with partially and completely unknown dynamics subject to variable sampled and delay output measurement[J]. Neurocomputing, 2023, 561: 126865. doi: 10.1016/j.neucom.2023.126865.
[17]	吴阳, 张建成. 同时含有未知输入和测量干扰系统全维和降维观测器设计[J]. 自动化学报, 2022, 48(8): 2108–2118. doi: 10.16383/j.aas.c190505. WU Yang and ZHANG Jiancheng. Full-and reduced-order observer design for systems with both the unknown inputs and measurement disturbances[J]. Acta Automatica Sinica, 2022, 48(8): 2108–2118. doi: 10.16383/j.aas.c190505.
[18]	CUI Wenqi, JIANG Yan, and ZHANG Baosen. Reinforcement learning for optimal primary frequency control: A Lyapunov approach[J]. IEEE Transactions on Power Systems, 2023, 38(2): 1676–1688. doi: 10.1109/TPWRS.2022.3176525.
[19]	BARBALHO P I N, LACERDA V A, FERNANDES R A S, et al. Deep reinforcement learning-based secondary control for microgrids in islanded mode[J]. Electric Power Systems Research, 2022, 212: 108315. doi: 10.1016/j.jpgr.2022.108315.
[20]	范培潇, 柯松, 杨军, 等. 基于改进多智能体深度确定性策略梯度的多微网负荷频率协同控制策略[J]. 电网技术, 2022, 46(9): 3504–3514. doi: 10.13335/j.1000-3673.pst.2021.1918. FAN Peixiao, KE Song, YANG Jun, et al. Load frequency coordinated control strategy of multi-microgrid based on improved MA-DDPG[J]. Power System Technology, 2022, 46(9): 3504–3514. doi: 10.13335/j.1000-3673.pst.2021.1918.
[21]	李峰, 王琦, 胡健雄, 等. 数据与知识联合驱动方法研究进展及其在电力系统中应用展望[J]. 中国电机工程学报, 2021, 41(13): 4377–4389. doi: 10.13334/j.0258-8013.pcsee.202468. LI Feng, WANG Qi, HU Jianxiong, et al. Combined data-driven and knowledge-driven methodology research advances and its applied prospect in power systems[J]. Proceedings of the CSEE, 2021, 41(13): 4377–4389. doi: 10.13334/j.0258-8013.pcsee.202468.
[22]	胡子剑, 高晓光, 万开方, 等. 异策略深度强化学习中的经验回放研究综述[J]. 自动化学报, 2023, 49(11): 2237–2256. doi: 10.16383/j.aas.c220648. HU Zijian, GAO Xiaoguang, WAN Kaifang, et al. Research on experience replay of off-policy deep reinforcement learning: a review[J]. Acta Automatica Sinica, 2023, 49(11): 2237–2256. doi: 10.16383/j.aas.c220648.
[23]	ZHAO Fuyu, GAO Weinan, LIU Tengfei, et al. Adaptive optimal output regulation of linear discrete-time systems based on event-triggered output-feedback[J]. Automatica, 2022, 137: 110103. doi: 10.1016/j.automatica.2021.110103.
[24]	CHENG Zihao, HU Songlin, YUE Dong, et al. Interval secure event-triggered mechanism for load frequency control active defense against DoS attack[J]. IEEE Transactions on Cybernetics, 2025, 55(2): 981–994. doi: 10.1109/TCYB.2024.3488208.
[25]	ZHANG Xianming, HAN Qinglong, GE Xiaohua, et al. An overview of recent advances in event-triggered control[J]. Science China Information Sciences, 2025, 68(6): 161201. doi: 10.1007/s11432-024-4437-9. (查阅网上资料,本条文献与第14条文献重复,请确认).
[26]	HASSANI H, NIKAN S, and SHAMI A. Traffic navigation via reinforcement learning with episodic-guided prioritized experience replay[J]. Engineering Applications of Artificial Intelligence, 2024, 137: 109147. doi: 10.1016/j.engappai.2024.109147.
[27]	ZHU Guibing, MA Yong, and HU Songlin. Event-triggered adaptive PID fault-tolerant control of underactuated ASVs under saturation constraint[J]. IEEE Transactions on Systems, Man, and Cybernetics: Systems, 2023, 53(8): 4922–4933. doi: 10.1109/TSMC.2023.3256538.
[28]	DELIMPALTADAKIS G, LAURENTI L, and MAZO M. Formal analysis of the sampling behavior of stochastic event-triggered control[J]. IEEE Transactions on Automatic Control, 2024, 69(7): 4491–4505. doi: 10.1109/TAC.2023.3333748.
[29]	陈红松, 刘新蕊, 陶子美, 等. 基于深度学习的时序数据异常检测研究综述[J]. 信息网络安全, 2025, 25(3): 364–391. doi: 10.3969/j.issn.1671-1122.2025.03.002. CHEN Hongsong, LIU Xinrui, TAO Zimei, et al. A survey of anomaly detection model for time series data based on deep learning[J]. Netinfo Security, 2025, 25(3): 364–391. doi: 10.3969/j.issn.1671-1122.2025.03.002.