Skip to main navigation Skip to search Skip to main content

An Ergodic CuSum Algorithm for False Data Injection Attacks Detection in DC Microgrids

  • SUNY Buffalo

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

DC microgrids have widely adopted hierarchical control architecture through distributed generation units (DGUs) to enhance reliability and scalability. However, this makes the system vulnerable to false data injection attacks (FDIAs), which can disrupt system stability or shift the operating point. While observers are commonly used to detect FDIAs, some FDIAs can be stealthy, or observers lack sufficient sensitivity for reliable identification. To address this, we propose a quickest change detection (QCD) method based on an unknown input observer (UIO) estimation error model to detect the FDIAs that are stealthy to the UIOs. The Ergodic CuSum algorithm is designed and can be efficiently updated using estimation error observations. The approach is validated through Simulink and real-time simulations.

Original languageEnglish
Title of host publication2025 IEEE Energy Conversion Conference Congress and Exposition, ECCE 2025
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9798331541309
DOIs
StatePublished - 2025
Event17th Annual IEEE Energy Conversion Conference Congress and Exposition, ECCE 2025 - Philadelphia, United States
Duration: Oct 19 2025Oct 23 2025

Publication series

Name2025 IEEE Energy Conversion Conference Congress and Exposition, ECCE 2025

Conference

Conference17th Annual IEEE Energy Conversion Conference Congress and Exposition, ECCE 2025
Country/TerritoryUnited States
CityPhiladelphia
Period10/19/2510/23/25

Keywords

  • Attack
  • CuSum
  • Detection
  • Microgrid
  • Observer

Fingerprint

Dive into the research topics of 'An Ergodic CuSum Algorithm for False Data Injection Attacks Detection in DC Microgrids'. Together they form a unique fingerprint.

Cite this