International Conference on Quantitative Evaluation of Systems

QEST 2015: Quantitative Evaluation of Systems pp 70-85

PCA-Based Method for Detecting Integrity Attacks on Advanced Metering Infrastructure

  • Varun Badrinath Krishna
  • Gabriel A. Weaver
  • William H. Sanders
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 9259)

Abstract

Electric utilities are in the process of installing millions of smart meters around the world, to help improve their power delivery service. Although many of these meters come equipped with encrypted communications, they may potentially be vulnerable to cyber intrusion attempts. These attempts may be aimed at stealing electricity, or destabilizing the electricity market system. Therefore, there is a need for an additional layer of verification to detect these intrusion attempts. In this paper, we propose an anomaly detection method that uniquely combines Principal Component Analysis (PCA) and Density-Based Spatial Clustering of Applications with Noise (DBSCAN) to verify the integrity of the smart meter measurements. Anomalies are deviations from the normal electricity consumption behavior. This behavior is modeled using a large, open database of smart meter readings obtained from a real deployment. We provide quantitative arguments that describe design choices for this method and use false-data injections to quantitatively compare this method with another method described in related work.

Keywords

Smart Meter Grid Anomaly Detection Principal Component Analysis Data Cyber-physical AMI PCA SVD DBSCAN Electricity Theft Energy Computer Communication Network Security 

Copyright information

© Springer International Publishing Switzerland 2015

Authors and Affiliations

  • Varun Badrinath Krishna
    • 1
  • Gabriel A. Weaver
    • 1
  • William H. Sanders
    • 1
  1. 1.Information Trust Institute, Department of Electrical and Computer EngineeringUniversity of Illinois at Urbana-ChampaignUrbanaUSA

Personalised recommendations