Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Machine-learned Adversarial Attacks against Fault Prediction Systems in Smart Electrical Grids

Carmelo Ardito, Yashar Deldjoo, Tommaso Di Noia, Eugenio Di Sciascio, Fatemeh Nazary, Giovanni Servedio

TL;DR

It is demonstrated first that the deep neural network method used in the smart grid is susceptible to adversarial perturbation and then how studies on fault localization and type classification illustrate the weaknesses of present ML algorithms in smart grids to various adversarial attacks.

Abstract

In smart electrical grids, fault detection tasks may have a high impact on society due to their economic and critical implications. In the recent years, numerous smart grid applications, such as defect detection and load forecasting, have embraced data-driven methodologies. The purpose of this study is to investigate the challenges associated with the security of machine learning (ML) applications in the smart grid scenario. Indeed, the robustness and security of these data-driven algorithms have not been extensively studied in relation to all power grid applications. We demonstrate first that the deep neural network method used in the smart grid is susceptible to adversarial perturbation. Then, we highlight how studies on fault localization and type classification illustrate the weaknesses of present ML algorithms in smart grids to various adversarial attacks

Machine-learned Adversarial Attacks against Fault Prediction Systems in Smart Electrical Grids

TL;DR

It is demonstrated first that the deep neural network method used in the smart grid is susceptible to adversarial perturbation and then how studies on fault localization and type classification illustrate the weaknesses of present ML algorithms in smart grids to various adversarial attacks.

Abstract

In smart electrical grids, fault detection tasks may have a high impact on society due to their economic and critical implications. In the recent years, numerous smart grid applications, such as defect detection and load forecasting, have embraced data-driven methodologies. The purpose of this study is to investigate the challenges associated with the security of machine learning (ML) applications in the smart grid scenario. Indeed, the robustness and security of these data-driven algorithms have not been extensively studied in relation to all power grid applications. We demonstrate first that the deep neural network method used in the smart grid is susceptible to adversarial perturbation. Then, we highlight how studies on fault localization and type classification illustrate the weaknesses of present ML algorithms in smart grids to various adversarial attacks
Paper Structure (9 sections, 5 equations, 2 figures, 1 table)

This paper contains 9 sections, 5 equations, 2 figures, 1 table.

Table of Contents

  1. Introduction and Context
  2. Approach
  3. Experimental Evaluation
  4. Datasets
  5. Adversarial Attacks
  6. Experiments and Results
  7. Explored Machine-Learnings Tasks
  8. Results
  9. Conclusion

Figures (2)

  • Figure 1: A hypothetical illustration of targeted adversarial attacks against fault zone prediction in smart grids. As a result of an adversarial attack on a fault location prediction system, dispatch recovery groups were dispatched to zone 3 by accident rather than zone 2, where they belonged.
  • Figure 2: Three tasks under targeted and untargeted adversarial attacks. Classification accuracy for $FZC = 0.7134$, $FTC = 0.4569$, and $FZC + FTC = 0.4543$. Best results for C&W were obtained under $\ell_{\infty}$ for untargeted attacks and $\ell_{2}$ for targeted attacks. Note that the starting point of noise power for all attacks and random noise is $0.001$.

Theorems & Definitions (2)

  • definition 1: Targeted adversarial attack
  • definition 2: Untargeted attack