Skip to content
Preprint

Detect, Repair and Audit: A Hierarchical Multi-Agent Defense Against Adversarial Attacks in Energy Forecasting

Nov 2026 · 0 citations · 42 references
Computer Science

TL;DR

A Multi-Agent Reinforcement Learning (MARL) detector coupled with signal repair at local control centers, designed to mitigate distributional data drift through complementary feature-based detection and clean-reference-calibrated repair while meeting latency budgets, is developed.

Abstract

Accurate electricity-demand forecasting is essential for the modern energy infrastructure. Recently, Times-Series Large Language Models (TS-LLMs) have emerged as a promising approach to forecasting more accurate demand from historical consumption data in zero- and few-shot settings, helping energy operators effectively plan supply and allocate energy resources. However, these models are vulnerable to compromised sensor inputs that can distort forecasts and lead to erroneous market orders or unnecessary load shedding. Existing defenses focus primarily on end-device forecasting, overlooking the distinct data-drift, latency, and audit requirements across energy infrastructure. To address this gap, we extract from major energy standards and regulatory frameworks a three-layer conceptual model that captures the security postures for each layer. Guided by this model, we develop a Multi-Agent Reinforcement Learning (MARL) detector coupled with signal repair at local control centers, designed to mitigate distributional data drift through complementary feature-based detection and clean-reference-calibrated repair while meeting latency budgets. We further deploy an evidence-grounded LLM auditing agent with human-in-the-loop validation for audit-critical Virtual Power Plants (VPPs). In our evaluation, the best MARL configuration achieves an F1 score above 0.91 across all dataset-attack pairs, while signal repair reduces adversarial-to-clean input Relative Squared Error (RSE) by up to 83%. The few-shot auditor detects 77.78% of attacks in the Sherlock dataset. On the evaluated hardware, detection and repair require 7.95 ms on average, with a worst-observed latency of 14.50 ms, remaining within the local-center latency budget informed by IEEE 1646.

View source

Similar papers

Open access Nov 2026

A lightweight closed-loop adaptive intrusion detection framework for cyber resilience in critical infrastructure

Most operators of energy, health, or transport facilities do not have at their disposal the kind of computational resources that large-scale deep-learning detectors require. Being energy and health care and transportation facility operators, they seek an intrusion detection approach that is responsive to environmen...

M. Al Ghamdi, Ibrahim Alghamdi, Rami J. Alzahrani · 0 citations
Open access Oct 2026

Robust uncertainty-aware attack-fault diagnosis in hydrogen-based energy system based on a copula-meta-learning approach

Active Distribution Power Systems (ADPS), including hydrogen-based renewable energy units, are becoming essential components of modern electrical grids. These systems integrate distributed energy resources with advanced communication technology to enhance efficiency, flexibility, and reliability. However, a significant...

Narinderjit Singh Sawaran Singh, Alaa Hamid Mohammed, Z. A. Hussein et al. · 0 citations
Open access Nov 2026

Safe-ARCS: Risk-Budget Shielding for Hierarchical Reinforcement Learning in Automated Cybersecurity Incident Response

Automated incident response should contain malicious activity without imposing unacceptable disruption on legitimate services. We study this conflict in a reproducible six-step stochastic simulator with noisy alerts, five latent incident classes, immediate containment actions, and persistent strategic defenses. Safe-AR...

Asha Singh, Benjamin Zhang, Caroline Young · 0 citations
Preprint Oct 2026

ROAM: Reinforcement Learning Optimized Adaptive Moving Target Defense for Smart Grid

Smart grids rely on distributed phasor measurement units (PMUs) and wide-area communication networks whose data-acquisition pipelines are vulnerable to stealthy false data injection (FDI) attacks that evade conventional bad-data detection (BDD). Moving target defense (MTD) counters this exposure by continuously remappi...

Mohammad Zakaria Haider, Prabin Mali, Muhammad Nadeem et al. · 0 citations
Open access Oct 2026

RiskCal-APT: A Decision-Centric Framework for Calibrated Advanced Persistent Threat Detection and High-Risk Mitigation

Advanced Persistent Threats (APTs) are difficult to defend against because the attacks are long-lived, span several hosts, and rely on tools that closely resemble legitimate administrator activity. Most intrusion detection systems (IDSs) treat APT defense as an ordinary classification problem, without probability calib...

Reeta Mishra, Neelu Chaudhary · 0 citations
Open access 2026

Adaptive Defense Optimization Under Intelligent Data Poisoning Attacks in Industrial Control System

As industrial control systems (ICSs) 4.0 evolve into open and networked environments, they face a critical challenge in maintaining robust closed-loop operation under feedback streams that are asynchronous, unreliable, and subject to adversarial data poisoning. This paper proposes a detection-aware stochastic optimizat...

Xiao Cai, Yan-Bin Sun, Hui Lu et al. · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.