基于BiLSTM与自注意力机制生成对抗网络的GSA防护方法

doi:10.11930/j.issn.1004-9649.202311025

中国电力 ›› 2024, Vol. 57 ›› Issue (9): 61-70.DOI: 10.11930/j.issn.1004-9649.202311025

• 面向电力基础设施的跨域攻击威胁与防御 • 上一篇下一篇

基于BiLSTM与自注意力机制生成对抗网络的GSA防护方法

吴辉¹(), 邹子威²(), 肖丰明¹, 刘杰¹, 闵陈鹏¹, 夏卓群³()

1. 五凌电力有限公司，湖南长沙　410004
2. 中南大学计算机学院，湖南长沙　410083
3. 长沙理工大学计算机与通信工程学院，湖南长沙　410076

收稿日期:2023-11-07 接受日期:2024-03-06 出版日期:2024-09-28 发布日期:2024-09-23
作者简介:吴辉（1985—），男，硕士，高级工程师，从事工业控制系统网络安全和计算机监控技术研究，E-mail：373663730@qq.com
邹子威（1997—），男，博士研究生，从事智能电网攻击检测、电力系统可靠性分析研究，E-mail：ziweizou@csu.edu.cn
夏卓群（1977—），男，通信作者，博士，教授，从事智能电网安全研究，E-mail：xiazhuoqun@csust.edu.cn
基金资助:
国家自然科学基金资助项目（智能电网边缘计算数据安全防护研究，52177067）；国家自然科学基金重点资助项目（面向大电网的网络攻击智能识别与安全防控理论与方法，U1966207）。

Defense Method for Smart Grid GPS Spoofing Attack Based on BiLSTM and Self-attention Mechanism Generative Adversarial Network

Hui WU¹(), Ziwei ZOU²(), Fengming XIAO¹, Jie LIU¹, Chenpeng MIN¹, Zhuoqun XIA³()

1. Wuling Electric Power Co., Ltd., Changsha 410004, China
2. School of Computer Science and Engineering, Central South University, Changsha 410083, China
3. School of Computer and Communication Engineering, Changsha University of Science and Technology, Changsha 410076, China

Received:2023-11-07 Accepted:2024-03-06 Online:2024-09-28 Published:2024-09-23
Supported by:
This work is supported by National Natural Science Foundation of China (Research on Data Security Protection of Smart Grid Edge Computing, No.52177067); Key Project of National Natural Science Foundation of China (Intelligent Identification and Security Prevention and Control Theory and Method of Network Attacks for Large Power Grids, No.U1966207).

摘要/Abstract

摘要：

相量测量单元（phasor measurement unit，PMU）是智能电网的重要组成部分，能精准同步采集电力数据。由于PMU使用全球定位系统（global positioning system，GPS）提供时间同步参考，容易遭受GPS欺骗攻击（GPS spoofing attack，GSA），影响正常的数据采集。现有GSA防御方法的修复精度较低且需要额外的硬件成本。为了解决上述问题，提出一种基于双向长短期记忆网络与自注意力机制生成对抗网络的GSA防护方法。首先，提出一种改进的带梯度惩罚的Wasserstein生成对抗网络（Wasserstein generative adversarial network with gradient penalty，WGAN-GP）模型，重新设计原有生成器和判别器的网络架构，并在生成器和判别器中分别引入双向长短期记忆网络以及自注意力机制，提升模型的生成性能和鉴别能力。其次，基于所提出的WGAN-GP模型，构建了一种GSA防御模型，其包含攻击检测网络和数据修复网络2个模块，分别用于检测智能电网GSA和修复受损的PMU测量数据。最后，在IEEE-39总线系统中模拟GSA攻击，并在相应的数据集验证方法的有效性。结果表明，与现有方法对比，所提方法在大部分性能指标上取得了领先的性能。

关键词: PMU, 攻击检测, 数据修复, GSA, WGAN-GP模型

Abstract:

Phasor measurement unit (PMU) plays a crucial role in smart grids, enabling precise synchronized acquisition of electric power data. Due to the use of the global positioning system (GPS) for time synchronization, the PMU is vulnerable to GPS spoofing attack (GSA), which impacts the normal data acquisition. The existing GSA defense methods have low restoration accuracy and require additional hardware costs. To address the aforementioned issues, this paper proposes a GSA defense method based on bidirectional long short-term memory (BiLSTM) network and self-attention mechanism generative adversarial network. Firstly, an improved WGAN-GP model is proposed to redesign the network architecture of the generator and discriminator, and the BiLSTM network and self-attention mechanism are incorporated into the generator and discriminator to enhance the model's generative performance and discriminative ability. Secondly, based on the proposed WGAN-GP model, a GSA defense model is constructed, which includes two crucial modules: an attack detection network and a data restoration network that are employed to detect the smart grid GSA and repair the compromised PMU measurement data, respectively. Finally, We simulated GSA attacks in the IEEE-39 bus system and validated the effectiveness of the proposed method on the corresponding dataset. The results show that compared to existing methods, the proposed approach outperforms in most performance indicators.

Key words: PMU, attack detection, data repair, GSA, WGAN-GP model

吴辉, 邹子威, 肖丰明, 刘杰, 闵陈鹏, 夏卓群. 基于BiLSTM与自注意力机制生成对抗网络的GSA防护方法[J]. 中国电力, 2024, 57(9): 61-70.

Hui WU, Ziwei ZOU, Fengming XIAO, Jie LIU, Chenpeng MIN, Zhuoqun XIA. Defense Method for Smart Grid GPS Spoofing Attack Based on BiLSTM and Self-attention Mechanism Generative Adversarial Network[J]. Electric Power, 2024, 57(9): 61-70.

导出引用管理器 EndNote|Ris|BibTeX

链接本文: https://www.electricpower.com.cn/CN/10.11930/j.issn.1004-9649.202311025

https://www.electricpower.com.cn/CN/Y2024/V57/I9/61

图/表 13

图 1 突变型GSA示意

Fig.1 Schematic of mutant GSA

图 2 慢速持续型GSA示意

Fig.2 Schematic of slow sustained GSA

图 3 GSA防御模型

Fig.3 GSA defense model

图 4 数据修复网络训练流程

Fig.4 Training workflow of data repair network

图 5 生成器结构

Fig.5 Structure of the generator

图 6 判别器架构

Fig.6 Architecture of the discriminator

图 7 IEEE-39总线系统

Fig.7 IEEE-39 bus system

表 1 本文所用数据集

Table 1 Datasets used in this paper

类别	PMU个数	数据量
正常PMU数据	29	8759
遭受突变型GSA的PMU数据	4	1276
遭受慢速持续型GSA的PMU数据	6	1867

表 2 消融实验

Table 2 Ablation experiments

方法	准确率/%	精确率/%	召回率/%	F1 分数/%
WGAN-GP	87.23	82.06	80.36	81.56
WGAN-GP+自注意力层	91.64	84.07	81.97	83.64
WGAN-GP+BiLSTM	92.66	84.96	82.39	83.97
本文方法	95.24	86.15	81.63	85.07

表 3 不同算法中检测模型的指标对比情况

Table 3 Comparison of detection model metrics among different algorithms

算法	准确率/%	AUC/%	训练时间/s	检测时间/s
文献[16]	94.32	94.57	43.24	0.42
文献[18]	92.38	90.81	39.29	0.32
文献[17]	91.32	91.30	46.08	0.30
本文方法	95.24	95.58	47.98	0.50

图 8 单类型GSA场景中不同算法修复相似度对比

Fig.8 Comparison of repair similarity of different algorithms in single-type GSA scenarios

表 4 不同类型GSA场景中修复数据准确率

Table 4 The accuracy of repaired data in different types of GSA scenarios 单位： %

场景类型	本文方法	WGAN-GP模型	文献[19]	文献[22]
单类型GSA	94.44	92.38	85.68	93.74
多类型GSA	92.63	89.72	—	91.77

图 9 多类型GSA场景中不同算法修复相似度对比

Fig.9 Comparison of repair similarities of different algorithms in multi-type GSA scenarios

参考文献 32

1	CHAUHAN S V S, GAO G X. Synchrophasor data under GPS spoofing: attack detection and mitigation using residuals[J]. IEEE Transactions on Smart Grid, 2021, 12 (4): 3415- 3424. DOI
2	杨杰, 郭逸豪, 郭创新, 等. 考虑模型与数据双重驱动的电力信息物理系统动态安全防护研究综述[J]. 电力系统保护与控制, 2022, 50 (7): 176- 187.
	YANG Jie, GUO Yihao, GUO Chuangxin, et al. A review of dynamic security protection on a cyber physical power system considering model and data driving[J]. Power System Protection and Control, 2022, 50 (7): 176- 187.
3	XUE A C, XU F Y, XU J S, et al. Online pattern recognition and data correction of PMU data under GPS spoofing attack[J]. Journal of Modern Power Systems and Clean Energy, 2020, 8 (6): 1240- 1249. DOI
4	ALMUTAIRY F, SCEKIC L, MATAR M, et al. Detection and mitigation of GPS spoofing attacks on phasor measurement units using deep learning[J]. International Journal of Electrical Power & Energy Systems, 2023, 151, 109160.
5	SABOURI M, SIAMAK S, DEHGHANI M, et al. Increasing the resiliency of power systems in presence of GPS spoofing attacks: a data-driven deep-learning algorithm[J]. IET Generation, Transmission & Distribution, 2023, 17(20): 4525-4540.
6	刘依晗, 王宇飞. 新型电力系统中跨域连锁故障的演化机理与主动防御探索[J]. 中国电力, 2022, 55 (2): 62- 72, 81.
	LIU Yihan, WANG Yufei. Exploring the evolution mechanism and active defense of cross-domain cascading failures in new type power system[J]. Electric Power, 2022, 55 (2): 62- 72, 81.
7	韦磊, 徐江涛, 郭雅娟, 等. 基于信任机制的电力无线传感网络安全簇头选举算法[J]. 中国电力, 2023, 56 (8): 61- 67, 76.
	WEI Lei, XU Jiangtao, GUO Yajuan, et al. Trust mechanism based secure cluster head election algorithm for electric power wireless sensor networks[J]. Electric Power, 2023, 56 (8): 61- 67, 76.
8	SHEPARD D P, HUMPHREYS T E, FANSLER A A. Evaluation of the vulnerability of phasor measurement units to GPS spoofing attacks[J]. International Journal of Critical Infrastructure Protection, 2012, 5 (3/4): 146- 153.
9	王晓燕, 杨晶晶, 黄铭, 等. GNSS干扰和欺骗检测研究现状与展望[J/OL]. 信号处理: 1–21[2023-06-14].https://kns.cnki.net/kcms/detail/11.2406.TN.20230613.1343.014.html.
	WANG Xiaoyan, YANG Jingjing, HUANG Ming, et al. Research status and prospect of GNSS jamming and spoofing detection[J/OL]. Journal of Signal Processing: 1–21. [2023-06-14].https://kns.cnki.net/kcms/detail/11.2406.TN.20230613.1343.014.html.
10	LI H, WANG X. Detection of GPS spoofing through signal multipath signature analysis[C]// 2016 IEEE Canadian Conference on Electrical and Computer Engineering (CCECE). IEEE, 2016: 1–5.
11	WEN H, HUANG P Y R, DYER J, et al. Countermeasures for GPS signal spoofing[C]// Proceedings of the 18th international technical meeting of the satellite division of the institute of navigation (ION GNSS 2005). 2005: 1285–1290.
12	RANGANATHAN A, ÓLAFSDÓTTIR H, CAPKUN S. SPREE: a spoofing resistant GPS receiver[C]//Proceedings of the 22nd Annual International Conference on Mobile Computing and Networking. New York City, New York. ACM, 2016: 348–360.
13	GONG S P, ZHANG Z H, TRINKLE M, et al. GPS spoofing based time stamp attack on real time wide area monitoring in smart grid[C]//2012 IEEE Third International Conference on Smart Grid Communications (SmartGridComm). Tainan, Taiwan, China. IEEE, 2012: 300–305.
14	BHAMIDIPATI S, KIM K J, SUN H B, et al. GPS spoofing detection and mitigation in PMUs using distributed multiple directional antennas[C]//ICC 2019-2019 IEEE International Conference on Communications (ICC). Shanghai, China. IEEE, 2019: 1–7.
15	MAGIERA J, KATULSKI R. Detection and mitigation of GPS spoofing based on antenna array processing[J]. Journal of Applied Research and Technology, 2015, 13 (1): 45- 57. DOI
16	YANG L, LI Y, LI Z. Improved-ELM method for detecting false data attack in smart grid[J]. International Journal of Electrical Power & Energy Systems, 2017, 91, 183- 191.
17	WANG J X, TU W T, HUI L C K, et al. Detecting time synchronization attacks in cyber-physical systems with machine learning techniques[C]//2017 IEEE 37th International Conference on Distributed Computing Systems (ICDCS). Atlanta, GA, USA. IEEE, 2017: 2246–2251.
18	LI Y C, YANG S S. GPS Spoofing attack detection in smart grids based on improved CapsNet[J]. China Communications, 2021, 18 (3): 174- 186. DOI
19	FAN X Y, DU L, DUAN D L. Synchrophasor data correction under GPS spoofing attack: a state estimation-based approach[J]. IEEE Transactions on Smart Grid, 2018, 9 (5): 4538- 4546. DOI
20	RISBUD P, GATSIS N, TAHA A. Vulnerability analysis of smart grids to GPS spoofing[J]. IEEE Transactions on Smart Grid, 2018, 10 (4): 3535- 3548.
21	WU Z Y, ZORA L T, PHADKE A G. Simultaneous transmission line parameter and PMU measurement calibration[C]//2015 IEEE Power & Energy Society General Meeting. Denver, CO, USA. IEEE, 2015: 1–5.
22	李元诚, 杨珊珊. 基于改进自注意力机制生成对抗网络的智能电网GPS欺骗攻击防御方法[J]. 电力自动化设备, 2021, 41 (11): 100- 106.
	LI Yuancheng, YANG Shanshan. Defense method of smart grid GPS spoofing attack based on improved self-attention generative adversarial network[J]. Electric Power Automation Equipment, 2021, 41 (11): 100- 106.
23	RISBUD P, GATSIS N, TAHA A. Multi-period power system state estimation with PMUs under GPS spoofing attacks[J]. Journal of Modern Power Systems and Clean Energy, 2020, 8 (4): 597- 606. DOI
24	SABOURI M, SIAMAK S, DEHGHANI M, et al. Intelligent GPS spoofing attack detection in power grid[C]//2021 11th Smart Grid Conference (SGC). Tabriz, Iran, Islamic Republic of. IEEE, 2021: 1–6.
25	PAL S, SIKDAR B, CHOW J H. Classification and detection of PMU data manipulation attacks using transmission line parameters[J]. IEEE Transactions on Smart Grid, 2018, 9 (5): 5057- 5066. DOI
26	WU Y R, YIN H, QIU W, et al. Optimal PMU design based on sampling model and sensitivity analysis[J]. International Journal of Electrical Power & Energy Systems, 2023, 148, 109004.
27	XIA X F, LIN J, JIA Q N, et al. ETD-ConvLSTM: a deep learning approach for electricity theft detection in smart grids[J]. IEEE Transactions on Information Forensics and Security, 2023, 18, 2553- 2568. DOI
28	GREFF K, SRIVASTAVA R K, KOUTNIK J, et al. LSTM: a search space odyssey[J]. IEEE Trans Neural Netw Learn Syst, 2017, 28 (10): 2222- 2232. DOI
29	MOUNIR N, OUADI H, JRHILIFA I. Short-term electric load forecasting using an EMD-BI-LSTM approach for smart grid energy management system[J]. Energy and Buildings, 2023, 288, 113022. DOI
30	HE J L, ZHANG Z J, MA L R, et al. InFocus: amplifying critical feature influence on non-intrusive load monitoring through self-attention mechanisms[J]. IEEE Transactions on Smart Grid, 2023, 14 (5): 3828- 3840. DOI
31	LI Y, WEI X H, LI Y Z, et al. Detection of false data injection attacks in smart grid: a secure federated deep learning approach[J]. IEEE Transactions on Smart Grid, 2022, 13 (6): 4862- 4872. DOI
32	SUN C H, ZHANG X H, MENG H Y, et al. AC-WGAN-GP: generating labeled samples for improving hyperspectral image classification with small-samples[J]. Remote Sensing, 2022, 14 (19): 4910. DOI

[1]	周生存, 罗毅, 易煊承, 吴亚宁, 李丁, 熊逸. 考虑数据缺失的图注意力网络暂态稳定评估[J]. 中国电力, 2024, 57(5): 157-167.
[2]	夏小飞, 芦宇峰, 苏毅, 杨健. 基于相空间重构与改进GSA-SVM的高压断路器机械故障诊断[J]. 中国电力, 2021, 54(10): 169-176.
[3]	黄潇潇, 杨韬, 郑骁麟, 郭凌旭, 林济铿. 基于PMU量测信息的短线路同杆并架双回线参数辨识[J]. 中国电力, 2020, 53(7): 141-148.

基于BiLSTM与自注意力机制生成对抗网络的GSA防护方法

Defense Method for Smart Grid GPS Spoofing Attack Based on BiLSTM and Self-attention Mechanism Generative Adversarial Network

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 13

参考文献 32

相关文章 3

编辑推荐

Metrics

模态框（Modal）标题

基于BiLSTM与自注意力机制生成对抗网络的GSA防护方法

Defense Method for Smart Grid GPS Spoofing Attack Based on BiLSTM and Self-attention Mechanism Generative Adversarial Network

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 13

参考文献 32

相关文章 3

编辑推荐

Metrics