False Data Injection Attack Detection in Power Systems Based on CNN-BiLSTM-Attention

GAO Zhirong; YANG Shan; YU Xi; LUO Zhaoxu

doi:10.20204/j.sp.2025.04013

English Version Reader LoginJournal Login

您当前的位置：

首页 >

文章列表页 >

False Data Injection Attack Detection in Power Systems Based on CNN-BiLSTM-Attention

更新时间：2025-10-22

- False Data Injection Attack Detection in Power Systems Based on CNN-BiLSTM-Attention
- Vol. 53, Issue 4, Pages: 103-111(2025)
- 作者机构：
  
  湖南工业大学电气与信息工程学院,湖南,株洲,412007
- 作者简介：
- 基金信息：
- DOI：10.20204/j.sp.2025.04013
  CLC：
- Published：2025
- 稿件说明：
移动端阅览
高芷蓉,杨杉,喻希,罗朝旭.基于CNN-BiLSTM-Attention的电力系统虚假数据注入攻击检测[J].智慧电力,2025,53(4):103-111.

doi:10.20204/j.sp.2025.04013
高芷蓉,杨杉,喻希,罗朝旭.基于CNN-BiLSTM-Attention的电力系统虚假数据注入攻击检测[J].智慧电力,2025,53(4):103-111. DOI： 10.20204/j.sp.2025.04013.

doi:10.20204/j.sp.2025.04013 DOI：

摘要

虚假数据注入攻击（FDIA）检测对于电力系统的安全运行至关重要。为进一步提高FDIA检测方法的准确性，提出了一种引入注意力（Attention）机制的卷积神经网络和双向长短期记忆网络（CNN-BiLSTM）混合的检测方法。首先，利用CNN-BiLSTM网络充分挖掘FDIA的时空特征，以及特征之间的双向时序关系，提高模型的学习能力。其次，引入注意力机制对FDIA的关键特征信息进行聚焦，有效过滤冗余信息，进一步提高模型的准确性。最后，在IEEE 14总线测试系统上进行仿真验证，算例结果表明该方法具有强大的数据特征提取能力，能有效提高FDIA的检测精度。

Abstract

False data injection attack （FDIA） detection is critical for the secure operation of power systems. To further improve the accuracy of FDIA detection methods， this paper proposes a hybrid detection method combining a convolutional neural network （CNN） and bidirectional long short-term memory network （BiLSTM） integrated with an attention mechanism. First， the CNN-BiLSTM network was utilized to fully extract the spatiotemporal features of FDIA and bidirectional temporal relationships among features， thereby enhancing the model’s learning capability. Second， the attention mechanism was introduced to focus on the key feature information of FDIA， effectively filtering redundant information to further improve the model’s accuracy. Finally， simulations were conducted on the IEEE 14-bus test system. Case study results demonstrate that the proposed method exhibits robust data feature extraction capabilities and significantly enhances FDIA detection precision.

关键词

Keywords

references

黄雨涵，丁涛，李雨婷，等.碳中和背景下能源低碳化技术综述及对新型电力系统发展的启示[J].中国电机工程学报，2021，41（S1）: 28-51.

周劼英，张晓，邵立嵩，等.新型电力系统网络安全防护挑战与展望[J].电力系统自动化，2023，47（8）: 15-24.

陈柏任，夏候凯顺，李梦诗. 基于数据驱动的电力系统虚假数据注入攻击防御框架的研究[J]. 电测与仪表，2024，61（12）：10-16.

LI B，XIAO G，LU R，et al.On feasibility and limitations of detecting false data injection attacks on power grid state estimation using D-FACTS devices[J].IEEE Transactions on Industrial Informatics，2020，16（2）: 854-864.

LI X，WANG Z，ZHANG C，et al.A novel dynamic watermarking-based EKF detection method for FDIAs in smart grid[J].IEEE/CAA Journal of Automatica Sinica，2022，9（7）: 1319-1322.

GUO L，SUN R，HAN X.FDIA system for sensors of the aero-engine control system based on the immune fusion Kalman filter[J].Mathematical Problems in Engineering，2021（1）: 1-17.

张佳伟，张江南，吴坡，等.基于融合容积卡尔曼滤波的直流微电网虚假数据检测[J].智慧电力，2024，52（3）: 87-93.

杨杉，谭博，郭静波.基于双马尔科夫链的新型能源互联网虚假数据注入攻击检测[J].电力自动化设备，2021，41（2）: 131-137.

MUSLEH A，CHEN G，DONG Z.A survey on the detection algorithms for false data injection attacks in smart grids[J].IEEE Transactions on Smart Grid，2020，11（3）: 2218-2234.

刘增稷，王琦，薛彤，等.电力系统中数据驱动算法安全威胁分析及应对方法研究[J].中国电机工程学报，2023，43（12）: 4538-4554.

姜竹青，门爱东，王海婴.计算机视觉中的深度学习[M].北京: 电子工业出版社，2021.

PAN H，FENG X，NA C，et al.A model for detecting false data injection attacks in smart grids based on the method utilized for image coding[J].IEEE Systems Journal，2023，17（4）: 6181-6191.

李元诚，曾婧.基于改进卷积神经网络的电网假数据注入攻击检测方法[J].电力系统自动化，2019，43（20）: 97-104.

YU B，WANG Z，LIU S，et al.The data dimensionality reduction and bad data detection in the process of smart grid reconstruction through machine learning[J].PLoS ONE，2020，15（10）: e0237994.

SHEN K，YAN W，NI H，et al.Localization of false data injection attack in smart grids based on SSA-CNN[J].Information，2023，14（3）: 180.

SUN K，QIU W，YAO W，et al.Frequency injection based HVDC attack-defense control via squeeze-excitation double CNN[J].IEEE Transactions on Power Systems，2021，PP（99）: 1-1.

XIA W，HE D，YU L.Locational detection of false data injection attacks in smart grids: A graph convolutional attention network approach[J].IEEE Internet of Things Journal，2024，11（6）: 9324-9337.

ZHU J，GAO W，LI Y，et al.Power System State Estimation Based on Fusion of PMU and SCADA Data[J].Energies，2024，17（11）: 2609.

杨玉泽，刘文霞，李承泽，等.面向电力SCADA系统的FDIA检测方法综述[J].中国电机工程学报，2023，43（22）: 8602-8622.

闫照康，马刚，冯瑞，等. 基于改进LSTM算法的综合能源系统多元负荷预测[J]. 分布式能源，2024，9（2）：30-38.

凌佳凯，章逸舟，胡金峰，等. 基于CNN-LSTM-Attention的配电网拓扑实时辨识方法[J]. 浙江电力，2024，43（3）：84-94.

苑开波，罗萍萍，王高猛，等.基于灰色关联分析法的电力系统隐蔽性数据攻击检测新方法[J].电工电能新技术，2019，38（1）: 17-23.

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Fault Classification in Distribution Network Based on CNN-SVM

A Review of Dynamic Characteristic Analysis Methods for Low Frequency Oscillation Modes in Power Systems

An Overview of Delayed Voltage Recovery in Power Systems

A Novel SVC Fuzzy Logic Controller to Improve the Transient Stability of Power System

New Approach to Calculate Load Margin Considering N-1 Static Voltage Stability Constraints of Power System

Related Author

JI Xingquan1

CHEN Jinshuo1

ZHANG Yumin1

LIU Qi2

GONG Zheng3

XU Bo4

LI Xun

GONG Qing-wu

Related Institution

State Grid Energy Research Institute Co.， Ltd.

State Grid Weifang Power Supply Company

Key Laboratory of Smart Grid of Ministry of Education， Tianjin University

1. College of Electrical Engineering and Automation， Shandong University of Science and Technology

School of Electrical Engineering， Xi’an Jiaotong University

AI问答

⁰