Abstract :

The Increasing Integration Of Smart Grid Technologies Has Significantly Improved The Efficiency, Reliability, And Automation Of Modern Power Systems. However, This Advancement Has Also Introduced Critical Cybersecurity Vulnerabilities, Particularly In The Form Of False Data Injection (FDI) Attacks, Which Can Compromise System Stability And Decisionmaking Processes. To Address These Challenges, This Paper Proposes An Efficient Privacyenhancing Cross-silo Federated Learning Framework For Detecting FDI Attacks In Smart Grids. The Proposed Approach Enables Multiple Distributed Entities, Such As Substations And Control Centers, To Collaboratively Train Machine Learning Models Without Sharing Raw Data, Thereby Preserving Data Privacy And Security. The Framework Incorporates Advanced Privacypreserving Techniques, Including Secure Aggregation And Differential Privacy, To Ensure That Sensitive Information Remains Protected During The Training Process. By Leveraging Crosssilo Federated Learning, The System Effectively Utilizes Decentralized Data Sources To Improve Detection Accuracy While Maintaining Compliance With Data Protection Requirements. The Model Is Trained On Heterogeneous Datasets Representing Different Grid Environments, Enabling It To Generalize Effectively Across Diverse Attack Scenarios. Experimental Results Demonstrate That The Proposed System Achieves High Detection Accuracy, Reduced Communication Overhead, And Enhanced Robustness Against Adversarial Manipulation Compared To Traditional Centralized Approaches. Additionally, The Framework Ensures Scalability And Adaptability, Making It Suitable For Real-world Smart Grid Deployments. Overall, This Work Contributes To Strengthening Smart Grid Cybersecurity By Combining Federated Learning With Privacy-enhancing Mechanisms For Reliable And Efficient FDI Attack Detection.

Published:

12-5-2026

Issue:

Vol. 26 No. 5 (2026)

Page Nos:

1099-1105

Section:

Articles

License:

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

How to Cite