Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Uncertainty-Aware Federated Learning for Cyber-Resilient Microgrid Energy Management

Oluleke Babayomi, Dong-Seong Kim

TL;DR

The paper tackles the challenge of maintaining economic efficiency and reliability of microgrid EMS under cyberattacks that corrupt PV forecasts. It introduces a cyber-resilient framework that fuses Federated LSTM PV forecasting with a two-stage cascade anomaly detector and Monte Carlo Dropout-based uncertainty quantification, followed by data sanitization and EMS optimization. The results show the cascade detector reduces false positives by about 70%, recovers 93.7% of forecast degradation, and achieves around 5% operational cost savings, mitigating roughly 34.7% of attack-induced losses. This work demonstrates that precision-focused, multi-signal defense can preserve forecast integrity and economic performance in decentralized microgrids, enabling trustworthy federated learning in critical energy infrastructure.

Abstract

Maintaining economic efficiency and operational reliability in microgrid energy management systems under cyberattack conditions remains challenging. Most approaches assume non-anomalous measurements, make predictions with unquantified uncertainties, and do not mitigate malicious attacks on renewable forecasts for energy management optimization. This paper presents a comprehensive cyber-resilient framework integrating federated Long Short-Term Memory-based photovoltaic forecasting with a novel two-stage cascade false data injection attack detection and energy management system optimization. The approach combines autoencoder reconstruction error with prediction uncertainty quantification to enable attack-resilient energy storage scheduling while preserving data privacy. Extreme false data attack conditions were studied that caused 58% forecast degradation and 16.9\% operational cost increases. The proposed integrated framework reduced false positive detections by 70%, recovered 93.7% of forecasting performance losses, and achieved 5\% operational cost savings, mitigating 34.7% of attack-induced economic losses. Results demonstrate that precision-focused cascade detection with multi-signal fusion outperforms single-signal approaches, validating security-performance synergy for decentralized microgrids.

Uncertainty-Aware Federated Learning for Cyber-Resilient Microgrid Energy Management

TL;DR

The paper tackles the challenge of maintaining economic efficiency and reliability of microgrid EMS under cyberattacks that corrupt PV forecasts. It introduces a cyber-resilient framework that fuses Federated LSTM PV forecasting with a two-stage cascade anomaly detector and Monte Carlo Dropout-based uncertainty quantification, followed by data sanitization and EMS optimization. The results show the cascade detector reduces false positives by about 70%, recovers 93.7% of forecast degradation, and achieves around 5% operational cost savings, mitigating roughly 34.7% of attack-induced losses. This work demonstrates that precision-focused, multi-signal defense can preserve forecast integrity and economic performance in decentralized microgrids, enabling trustworthy federated learning in critical energy infrastructure.

Abstract

Maintaining economic efficiency and operational reliability in microgrid energy management systems under cyberattack conditions remains challenging. Most approaches assume non-anomalous measurements, make predictions with unquantified uncertainties, and do not mitigate malicious attacks on renewable forecasts for energy management optimization. This paper presents a comprehensive cyber-resilient framework integrating federated Long Short-Term Memory-based photovoltaic forecasting with a novel two-stage cascade false data injection attack detection and energy management system optimization. The approach combines autoencoder reconstruction error with prediction uncertainty quantification to enable attack-resilient energy storage scheduling while preserving data privacy. Extreme false data attack conditions were studied that caused 58% forecast degradation and 16.9\% operational cost increases. The proposed integrated framework reduced false positive detections by 70%, recovered 93.7% of forecasting performance losses, and achieved 5\% operational cost savings, mitigating 34.7% of attack-induced economic losses. Results demonstrate that precision-focused cascade detection with multi-signal fusion outperforms single-signal approaches, validating security-performance synergy for decentralized microgrids.

Paper Structure

This paper contains 26 sections, 7 equations, 2 figures, 4 tables.

Table of Contents

  1. Introduction
  2. System Architecture and Threat Model
  3. Federated Learning Architecture
  4. Threat Model: False Data Injection Attacks
  5. Proposed Cyber-Resilient Framework
  6. Two-Stage Cascade Anomaly Detection
  7. Stage 1: Autoencoder-Based Reconstruction Error Detection
  8. Stage 2: Uncertainty Quantification Refinement
  9. Data Filtering and Sanitization
  10. Federated LSTM Training on Filtered Data
  11. Attack-Resilient Energy Management System Optimization
  12. Experimental Setup and Performance Metrics
  13. Dataset and Preprocessing
  14. Performance Metrics
  15. Results and Analysis
  16. ...and 11 more sections

Figures (2)

  • Figure 1: System diagram. (a) Overall system architecture. (b) Proposed federated learning approach to FDIA-resilient PV forecasting and energy management.
  • Figure 2: Plots of results.