Data-Driven Detection of False Data Injection Attacks in Smart Grids using Self-Attention Enabled Conditional GAN with Gradient Penalty
Keywords:
Conditional generative adversarial network, deep learning, False data injection attack, Gradient penalty, Intrusion detection, Self-attention mechanism, Smart grid securityAbstract
The increasing integration of advanced information and communication technologies in power systems has increased their vulnerability to cyberattacks. False Data Injection Attacks (FDIAs) are a concerning and widely encountered cyberattack. FDIAs pose a critical threat to the security and reliability of modern power systems by manipulating measurement data. Traditional state estimation techniques often fail to detect stealthy FDIAs, particularly in large-scale grids. This paper proposes a Self-Attention-enabled Conditional Generative Adversarial Network with Gradient Penalty (SACGAN-GP) for effective FDIA detection. The framework leverages a self-attention mechanism to capture long-range dependencies among state variables, enhancing feature representation and improving detection performance. The gradient penalty term ensures training stability and mitigates mode collapse, a common issue in standard GANs. SACGAN-GP can effectively utilize limited labelled attack data and learn robust representations of normal and anomalous patterns, without relying heavily on large, labelled datasets as required by supervised models. Experimental validation on IEEE 14-bus and 118-bus test systems demonstrates that the proposed model performs better than other methods. The proposed method achieves an accuracy of 97.06%, F1-score of 98.28%, and an AUC of 100% on the IEEE 14-bus system, while attaining 95.59% accuracy, 97.39% F1-score, and 100% AUC on the IEEE 118-bus system. Detection times remain under 0.52 seconds, confirming the method’s applicability for real-time scenarios. Furthermore, attention heatmaps generated by the model provide interpretable insights into the localized impacts of FDIAs, offering a promising direction for intelligent, secure power grid monitoring.
Downloads
References
G. Del Río Castro, M. C. González Fernández, and Á. Uruburu Colsa, “Unleashing the convergence amid digitalization and sustainability towards pursuing the Sustainable Development Goals (SDGs): A holistic review,” Jan. 20, 2021, Elsevier Ltd. doi: 10.1016/j.jclepro.2020.122204.
X. Wang, H. Zhu, X. Luo, and X. Guan, “Data-Driven-Based Detection and Localization Framework Against False Data Injection Attacks in DC Microgrids,” IEEE Internet Things J, vol. 12, no. 17, pp. 36079–36093, 2025, doi: 10.1109/JIOT.2025.3579915.
L. Xiao, H. Chen, S. Xu, Z. Lv, C. Wang, and Y. Xiao, “Reinforcement Learning-Based False Data Injection Attacks in Smart Grids,” IEEE Trans Industr Inform, vol. 21, no. 4, pp. 3475–3484, 2025, doi: 10.1109/TII.2025.3528571.
H. Feng, Y. Han, F. Si, and Q. Zhao, “Detection of False Data Injection Attacks in Cyber-Physical Power Systems: An Adaptive Adversarial Dual Autoencoder With Graph Representation Learning Approach,” IEEE Trans Instrum Meas, vol. 73, pp. 1–11, 2024, doi: 10.1109/TIM.2023.3331398.
J. Xie, A. Rahman, and W. Sun, “Bayesian GAN-Based False Data Injection Attack Detection in Active Distribution Grids with DERs,” IEEE Trans Smart Grid, vol. 15, no. 3, pp. 3223–3234, 2024, doi: 10.1109/TSG.2023.3337340.
Y. Liu, P. Ning, and M. K. Reiter, “False data injection attacks against state estimation in electric power grids,” in ACM Transactions on Information and System Security, May 2011. doi: 10.1145/1952982.1952995.
G. Chaojun, P. Jirutitijaroen, and M. Motani, “Detecting False Data Injection Attacks in AC State Estimation,” IEEE Trans Smart Grid, vol. 6, no. 5, pp. 2476–2483, 2015, doi: 10.1109/TSG.2015.2388545.
G. Cheng, Y. Lin, J. Zhao, and J. Yan, “A Highly Discriminative Detector Against False Data Injection Attacks in AC State Estimation,” IEEE Trans Smart Grid, vol. 13, no. 3, pp. 2318–2330, 2022, doi: 10.1109/TSG.2022.3141803.
N. Li, J. Zhang, D. Ma, and J. Ding, “Enhancing Detection of False Data Injection Attacks in Smart Grid Using Spectral Graph Neural Network,” IEEE Trans Industr Inform, vol. 21, no. 6, pp. 4543–4553, 2025, doi: 10.1109/TII.2025.3545044.
S. H. Mohammed et al., Dual-hybrid intrusion detection system to detect False Data Injection in smart grids, vol. 20, no. 1 January. 2025. doi: 10.1371/journal.pone.0316536.
A. Takiddin, M. Ismail, R. Atat, K. R. Davis, and E. Serpedin, “Robust Graph Autoencoder-Based Detection of False Data Injection Attacks Against Data Poisoning in Smart Grids,” IEEE Transactions on Artificial Intelligence, vol. 5, no. 3, pp. 1287–1301, 2024, doi: 10.1109/TAI.2023.3286831.
Y. Wang, Y. Zhou, J. Ma, and Q. jin, “A locational false data injection attack detection method in smart grid based on adversarial variational autoencoders [Formula presented],” Appl Soft Comput, vol. 151, no. July 2023, p. 111169, 2024, doi: 10.1016/j.asoc.2023.111169.
K. Li, F. Li, B. Wang, and M. Shan, “False data injection attack sample generation using an adversarial attention-diffusion model in smart grids,” AIMS Energy, vol. 12, no. 6, pp. 1271–1293, 2024, doi: 10.3934/ENERGY.2024058.
A. Shees, M. Tariq, and A. I. Sarwat, “Cybersecurity in Smart Grids: Detecting False Data Injection Attacks Utilizing Supervised Machine Learning Techniques,” Energies (Basel), vol. 17, no. 23, 2024, doi: 10.3390/en17235870.
Y. Lin and J. Wang, “Probabilistic Deep Autoencoder for Power System Measurement Outlier Detection and Reconstruction,” IEEE Trans Smart Grid, vol. 11, no. 2, pp. 1796–1798, 2020, doi: 10.1109/TSG.2019.2937043.
S. A. Foroutan and F. R. Salmasi, “Detection of false data injection attacks against state estimation in smart grids based on a mixture Gaussian distribution learning method,” IET Cyber-Physical Systems: Theory & Applications, vol. 2, no. 4, pp. 161–171, 2017, doi: 10.1049/iet-cps.2017.0013.
M. Ozay, I. Esnaola, F. T. Yarman Vural, S. R. Kulkarni, and H. V. Poor, “Machine Learning Methods for Attack Detection in the Smart Grid,” IEEE Trans Neural Netw Learn Syst, vol. 27, no. 8, pp. 1773–1786, 2016, doi: 10.1109/TNNLS.2015.2404803.
J. J. Q. Yu, Y. Hou, and V. O. K. Li, “Online False Data Injection Attack Detection with Wavelet Transform and Deep Neural Networks,” IEEE Trans Industr Inform, vol. 14, no. 7, pp. 3271–3280, 2018, doi: 10.1109/TII.2018.2825243.
G. Zhang, J. Li, O. Bamisile, D. Cai, W. Hu, and Q. Huang, “Spatio-Temporal Correlation-Based False Data Injection Attack Detection Using Deep Convolutional Neural Network,” IEEE Trans Smart Grid, vol. 13, no. 1, pp. 750–761, 2022, doi: 10.1109/TSG.2021.3109628.
S. Liu, C. Wu, and H. Zhu, “Topology-Aware Graph Neural Networks for Learning Feasible and Adaptive AC-OPF Solutions,” IEEE Transactions on Power Systems, vol. 38, no. 6, pp. 5660–5670, 2023, doi: 10.1109/TPWRS.2022.3230555.
Y. Li, X. Wei, Y. Li, Z. Dong, and M. Shahidehpour, “Detection of False Data Injection Attacks in Smart Grid: A Secure Federated Deep Learning Approach,” IEEE Trans Smart Grid, vol. 13, no. 6, pp. 4862–4872, 2022, doi: 10.1109/TSG.2022.3204796.
M. Abdel-Basset, N. Moustafa, and H. Hawash, “Privacy-Preserved Generative Network for Trustworthy Anomaly Detection in Smart Grids: A Federated Semisupervised Approach,” IEEE Trans Industr Inform, vol. 19, no. 1, pp. 995–1005, 2023, doi: 10.1109/TII.2022.3165869.
M. R. Uddin, R. Rahman, and D. C. Nguyen, “False Data Injection Attack Detection in Edge-based Smart Metering Networks with Federated Learning,” 2024, doi: 10.1109/CCNC54725.2025.10976228.
Y. Zhang, J. Wang, and B. Chen, “Detecting False Data Injection Attacks in Smart Grids: A Semi-Supervised Deep Learning Approach,” IEEE Trans Smart Grid, vol. 12, no. 1, pp. 623–634, 2021, doi: 10.1109/TSG.2020.3010510.
N. Costilla-Enriquez and Y. Weng, “Attack Power System State Estimation by Implicitly Learning the Underlying Models,” IEEE Trans Smart Grid, vol. 14, no. 1, pp. 649–662, 2023, doi: 10.1109/TSG.2022.3197770.
S. Aldhaheri, “Aldhaheri, Sahar, and Abeer Alhuzali. ‘SGAN-IDS: Self-attention-based generative adversarial network against intrusion detection systems.’ Sensors 23.18 (2023): 7796.,” 2023.
