Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A Multi-Scale Attention-Based Attack Diagnosis Mechanism for Parallel Cyber-Physical Attacks in Power Grids

Junhao Ren, Kai Zhao, Guangxiao Zhang, Xinghua Liu, Chao Zhai, Gaoxi Xiao

TL;DR

This work addresses the challenge of parallel cyber-physical attacks on power grids by integrating a learning-based attack localization module with a rigorous optimization-based diagnosis stage. The CGCA-AL framework leverages multi-scale attention (GAT-CNN with MHCA) to produce attack-probability priors, which guide a meta-mixed-integer programming (MMIP) formulation for Line State Identification (LSI) under PCPA. The approach provides sufficient conditions for reconstructing attacked-area measurements, formulates PCPA as MMIP, and shows that incorporating CGCA-AL priors improves solution quality and robustness on IEEE 30-bus and 118-bus test systems. Simulation results demonstrate high localization accuracy, reduced false alarms, and scalable performance, highlighting the framework’s practical potential for enhancing grid resilience against sophisticated cyber-physical threats.

Abstract

Parallel cyber--physical attacks (PCPA) can simultaneously damage physical transmission lines and disrupt measurement data transmission in power grids, severely impairing system situational awareness and attack diagnosis. This paper investigates the attack diagnosis problem for linearized AC/DC power flow models under PCPA, where physical attacks include not only line disconnections but also admittance modifications, such as those caused by compromised distributed flexible AC transmission system (D-FACTS) devices. To address this challenge, we propose a learning-assisted attack diagnosis framework based on meta--mixed-integer programming (MMIP), which integrates a convolutional graph cross-attention attack localization (CGCA-AL) model. First, sufficient conditions for measurement reconstruction are derived, enabling the recovery of unknown measurements in attacked areas using available measurements and network topology information. Based on these conditions, the attack diagnosis problem is formulated as an MMIP model. The proposed CGCA-AL employs a multi-scale attention mechanism to predict a probability distribution over potential physical attack locations, which is incorporated into the MMIP as informative objective coefficients. By solving the resulting MMIP, both the locations and magnitudes of physical attacks are optimally estimated, and system states are subsequently reconstructed. Simulation results on IEEE 30-bus and IEEE 118-bus test systems demonstrate the effectiveness, robustness, and scalability of the proposed attack diagnosis framework under complex PCPA scenarios.

A Multi-Scale Attention-Based Attack Diagnosis Mechanism for Parallel Cyber-Physical Attacks in Power Grids

TL;DR

This work addresses the challenge of parallel cyber-physical attacks on power grids by integrating a learning-based attack localization module with a rigorous optimization-based diagnosis stage. The CGCA-AL framework leverages multi-scale attention (GAT-CNN with MHCA) to produce attack-probability priors, which guide a meta-mixed-integer programming (MMIP) formulation for Line State Identification (LSI) under PCPA. The approach provides sufficient conditions for reconstructing attacked-area measurements, formulates PCPA as MMIP, and shows that incorporating CGCA-AL priors improves solution quality and robustness on IEEE 30-bus and 118-bus test systems. Simulation results demonstrate high localization accuracy, reduced false alarms, and scalable performance, highlighting the framework’s practical potential for enhancing grid resilience against sophisticated cyber-physical threats.

Abstract

Parallel cyber--physical attacks (PCPA) can simultaneously damage physical transmission lines and disrupt measurement data transmission in power grids, severely impairing system situational awareness and attack diagnosis. This paper investigates the attack diagnosis problem for linearized AC/DC power flow models under PCPA, where physical attacks include not only line disconnections but also admittance modifications, such as those caused by compromised distributed flexible AC transmission system (D-FACTS) devices. To address this challenge, we propose a learning-assisted attack diagnosis framework based on meta--mixed-integer programming (MMIP), which integrates a convolutional graph cross-attention attack localization (CGCA-AL) model. First, sufficient conditions for measurement reconstruction are derived, enabling the recovery of unknown measurements in attacked areas using available measurements and network topology information. Based on these conditions, the attack diagnosis problem is formulated as an MMIP model. The proposed CGCA-AL employs a multi-scale attention mechanism to predict a probability distribution over potential physical attack locations, which is incorporated into the MMIP as informative objective coefficients. By solving the resulting MMIP, both the locations and magnitudes of physical attacks are optimally estimated, and system states are subsequently reconstructed. Simulation results on IEEE 30-bus and IEEE 118-bus test systems demonstrate the effectiveness, robustness, and scalability of the proposed attack diagnosis framework under complex PCPA scenarios.

Paper Structure

This paper contains 33 sections, 3 theorems, 23 equations, 3 figures, 5 tables, 2 algorithms.

Table of Contents

  1. Introduction
  2. Problem Formulation
  3. Power grid model
  4. Attack model of PCPA
  5. State Analysis of Power Grids under PCPA
  6. Data reconstruction for the attacked area
  7. unknown phase voltage angle $\vec{\theta}'_H$
  8. unknown power injection $\vec{p}'_H$
  9. Attack Diagnosis Formulation for PCPA
  10. Algorithm Design for Attack Diagnosis
  11. GAT Module
  12. Graph attention mechanism
  13. Message passing
  14. MHCA Mechanism
  15. Architecture of the CGCA-AL algorithm
  16. ...and 18 more sections

Key Result

Lemma 1

$\operatorname{supp}(A(\vec{\theta}- \vec{\theta}')- \vec{\Delta}) \subseteq{\cal V}_{H}$.

Figures (3)

  • Figure 1: Architecture of power grid and attack scheme of PCPA.
  • Figure 2: Scheme of Attack diagnosis process for PCPA.
  • Figure 3: The structure of the CGCA-AL algorithm. GAT-CNN layer is used to extract the high dimensional features from local spatial and long-range correlations, respectively. Multi head cross-attention is utilized to enhanced the high dimensional features of ${\cal V}_H$.

Theorems & Definitions (7)

  • Lemma 1
  • Lemma 2
  • Lemma 3
  • Remark
  • proof
  • proof
  • proof