Smart grid false data injection detection through federated learning with deep learning models
Decision Science Letters · 2026 · 15 (2) · 357-370
DOI: 10.5267/j.dsl.2026.2.004 ↗Kurz gefasst
Fachartikel in der Zeitschrift Decision Science Letters, 2026. Verfasst von Raseel Alshamasi und 1 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 1 Literaturangaben.
Abstract (Originalsprache)
The security of smart grids is seriously threatened by false data injection (FDI) attacks. Falsified data is maliciously injected into the grid's measurement and control systems as part of these attacks, which might seriously disrupt the power supply and jeopardize system integrity. In the context of smart grids, it is also imperative to address the issue of consumer privacy and the protection of their sensitive data. The main objective of this work is to provide a collaborative framework based on federated learning to detect various FDI dangers while protecting SG's resources and privacy. We have implemented several technologies that provide a good solution in order to accomplish this goal. Using a dataset designed to replicate attacks on the power system environment, we used federated learning to locally train models using the data stored on the sensors. The best model should then be chosen by comparing the outcomes. These outcomes demonstrate the potential of our framework, which has used mixed models to repel attacks, short-circuit faults, and maintain lines with a 98% accuracy rate during the federated learning phase.
Zitieren
APA
Alshamasi, R., & Ibrahim, D. M. (2026). Smart grid false data injection detection through federated learning with deep learning models. Decision Science Letters, 15(2), 357-370. https://doi.org/10.5267/j.dsl.2026.2.004
Vancouver
Alshamasi R, Ibrahim DM. Smart grid false data injection detection through federated learning with deep learning models. Decision Science Letters. 2026;15(2):357-370. doi:10.5267/j.dsl.2026.2.004
BibTeX
@article{Alshamasi2026,
author = {Alshamasi, Raseel and Ibrahim, Dina M.},
title = {Smart grid false data injection detection through federated learning with deep learning models},
journal = {Decision Science Letters},
year = {2026},
volume = {15},
number = {2},
pages = {357--370},
doi = {10.5267/j.dsl.2026.2.004}
}
EUVIMED