Federated Learning for Enhancing Cybersecurity Resilience in Distributed Energy Systems
Bibliographic record
Abstract
Abstract This research investigates federated learning (FL) as a novel approach to strengthen cybersecurity resilience in distributed energy systems (DES), including substations, distributed energy resources (DERs), and grid control networks. Traditional centralized security models are inadequate for modern energy infrastructure due to privacy constraints, bandwidth limitations, and the vast scale of distributed assets [1]. A federated learning architecture was deployed across critical energy assets, where each node locally trained machine learning models on operational telemetry without sharing raw data. Only model parameters were transmitted to a central aggregator, preserving privacy while enabling collaborative threat detection. The system was evaluated through testbed validation and pilot deployment across 20+ geographically distributed nodes. Results demonstrated significant improvements over centralized approaches: 15% enhancement in anomaly detection accuracy, 100% threat detection rate with zero false positives after seven training rounds, 22% reduction in communication bandwidth requirements, and 86% F1-score maintenance across the distributed network. The edge agent required only 16 MB RAM and 25 MB disk space, enabling deployment on resource-constrained industrial devices. This framework pioneers the integration of federated learning into energy cybersecurity, providing a scalable, privacy-preserving solution that addresses current limitations while ensuring regulatory compliance with NERC CIP and IEC 62443 standards.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.004 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".