Consensus‐Based Ensemble Learning for False Data Injection Attack Detection Utilising Conceptual Dimensionality Reduction and Temporal Changes
IET Smart Grid · 2026-01 · 9 (1)
DOI: 10.1049/stg2.70109 ↗Kurz gefasst
Fachartikel in der Zeitschrift IET Smart Grid, online erschienen am 30.07.2026. Verfasst von Hassan Manafi und 2 weiteren Autor:innen. Die Arbeit nennt 33 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).
Abstract (Originalsprache)
The state estimator (SE) module plays an important role in power system monitoring through providing the best consistent states using the available redundant and noisy measurements. As the first module in the transmission control centres, the SE is more frequently targeted by cyberattacks, especially false data injection attacks (FDIAs). Various statistical and machine learning (ML)‐based methods have been proposed in the literature to detect FDIAs on the SE, often assuming simplifications regarding the calculation mechanism (DC formulation) and the attacker’s information level (complete knowledge). This paper proposes a new ML‐based detection approach for FDIAs on the AC SE with partial information about the attacker. Ensemble learning (EL) is utilised, considering the weighted voting consensus among individual ML‐based classifiers, that is, K‐nearest neighbours (KNNs), support vector machines (SVMs) and Gaussian naive Bayes (GNB). The proposed framework uses temporal changes across various time frames along with a measurement data diversification method to enhance the detection accuracy. Moreover, dimensionality reduction is performed based on suggested conceptual indices and the principal component analysis (PCA) algorithm. The desirable accuracy (over 99%) of the proposed method is demonstrated based on evaluation on the IEEE 118‐bus system while showing a better performance compared to the literature.
Zitieren
APA
Manafi, H., Delkhosh, H., & Seifi, H. (2026). Consensus‐Based Ensemble Learning for False Data Injection Attack Detection Utilising Conceptual Dimensionality Reduction and Temporal Changes. IET Smart Grid, 9(1). https://doi.org/10.1049/stg2.70109
Vancouver
Manafi H, Delkhosh H, Seifi H. Consensus‐Based Ensemble Learning for False Data Injection Attack Detection Utilising Conceptual Dimensionality Reduction and Temporal Changes. IET Smart Grid. 2026;9(1). doi:10.1049/stg2.70109
BibTeX
@article{Manafi2026,
author = {Manafi, Hassan and Delkhosh, Hamed and Seifi, Hossein},
title = {Consensus‐Based Ensemble Learning for False Data Injection Attack Detection Utilising Conceptual Dimensionality Reduction and Temporal Changes},
journal = {IET Smart Grid},
year = {2026},
volume = {9},
number = {1},
doi = {10.1049/stg2.70109}
}
EUVIMED