A two-layer game theoretical attack-defense model for a false data injection attack against power systems

Qi Wang*, Wei Tai, Yi Tang, Ming Ni, Shi You

*Corresponding author for this work

Research output: Contribution to journalJournal articleResearchpeer-review

1 Downloads (Pure)

Abstract

With the widespread application of information and communication technology in power systems, cyber security has become critical for ensuring effective measurement, communication, calculation and execution. A successful false data injection attack (FDIA) can bypass traditional systems for identifying bad data and interfere with decision making in control centers, thus causing power system failures. In this paper, the measurements within the phasor measurement units (PMUs) are used as attack and defense objects, and load shedding resulting from line failure is utilized to quantify the attack consequences. To construct the optimal defense against an FDIA, this paper examines the characteristics and construction of FDIAs from the perspective of attackers, aiming to reveal the shortcomings of traditional bad data identification. From the perspective of those defending against an FDIA, a two-layer defense model is proposed that includes detection and protection. First, information from multiple sources is utilized to improve the detection of false data. Second, extra PMUs are added as a protection method to strengthen the measurement redundancy. A zero-sum static game algorithm is applied to optimize the deployment of defense resources. The effectiveness of the proposed strategy is tested using IEEE 14-bus, 57-bus and 118-bus systems, and the impact of information asymmetry is also discussed.

Original languageEnglish
JournalInternational Journal of Electrical Power and Energy Systems
Volume104
Pages (from-to)169-177
ISSN0142-0615
DOIs
Publication statusPublished - 1 Jan 2019

Keywords

  • False data injection attack
  • Game theory
  • Optimal load shedding
  • State estimation

Cite this

@article{c54b8aea9f94481ca19cceb8504c624d,
title = "A two-layer game theoretical attack-defense model for a false data injection attack against power systems",
abstract = "With the widespread application of information and communication technology in power systems, cyber security has become critical for ensuring effective measurement, communication, calculation and execution. A successful false data injection attack (FDIA) can bypass traditional systems for identifying bad data and interfere with decision making in control centers, thus causing power system failures. In this paper, the measurements within the phasor measurement units (PMUs) are used as attack and defense objects, and load shedding resulting from line failure is utilized to quantify the attack consequences. To construct the optimal defense against an FDIA, this paper examines the characteristics and construction of FDIAs from the perspective of attackers, aiming to reveal the shortcomings of traditional bad data identification. From the perspective of those defending against an FDIA, a two-layer defense model is proposed that includes detection and protection. First, information from multiple sources is utilized to improve the detection of false data. Second, extra PMUs are added as a protection method to strengthen the measurement redundancy. A zero-sum static game algorithm is applied to optimize the deployment of defense resources. The effectiveness of the proposed strategy is tested using IEEE 14-bus, 57-bus and 118-bus systems, and the impact of information asymmetry is also discussed.",
keywords = "False data injection attack, Game theory, Optimal load shedding, State estimation",
author = "Qi Wang and Wei Tai and Yi Tang and Ming Ni and Shi You",
year = "2019",
month = "1",
day = "1",
doi = "10.1016/j.ijepes.2018.07.007",
language = "English",
volume = "104",
pages = "169--177",
journal = "International Journal of Electrical Power & Energy Systems",
issn = "0142-0615",
publisher = "Elsevier",

}

A two-layer game theoretical attack-defense model for a false data injection attack against power systems. / Wang, Qi; Tai, Wei; Tang, Yi; Ni, Ming; You, Shi.

In: International Journal of Electrical Power and Energy Systems, Vol. 104, 01.01.2019, p. 169-177.

Research output: Contribution to journalJournal articleResearchpeer-review

TY - JOUR

T1 - A two-layer game theoretical attack-defense model for a false data injection attack against power systems

AU - Wang, Qi

AU - Tai, Wei

AU - Tang, Yi

AU - Ni, Ming

AU - You, Shi

PY - 2019/1/1

Y1 - 2019/1/1

N2 - With the widespread application of information and communication technology in power systems, cyber security has become critical for ensuring effective measurement, communication, calculation and execution. A successful false data injection attack (FDIA) can bypass traditional systems for identifying bad data and interfere with decision making in control centers, thus causing power system failures. In this paper, the measurements within the phasor measurement units (PMUs) are used as attack and defense objects, and load shedding resulting from line failure is utilized to quantify the attack consequences. To construct the optimal defense against an FDIA, this paper examines the characteristics and construction of FDIAs from the perspective of attackers, aiming to reveal the shortcomings of traditional bad data identification. From the perspective of those defending against an FDIA, a two-layer defense model is proposed that includes detection and protection. First, information from multiple sources is utilized to improve the detection of false data. Second, extra PMUs are added as a protection method to strengthen the measurement redundancy. A zero-sum static game algorithm is applied to optimize the deployment of defense resources. The effectiveness of the proposed strategy is tested using IEEE 14-bus, 57-bus and 118-bus systems, and the impact of information asymmetry is also discussed.

AB - With the widespread application of information and communication technology in power systems, cyber security has become critical for ensuring effective measurement, communication, calculation and execution. A successful false data injection attack (FDIA) can bypass traditional systems for identifying bad data and interfere with decision making in control centers, thus causing power system failures. In this paper, the measurements within the phasor measurement units (PMUs) are used as attack and defense objects, and load shedding resulting from line failure is utilized to quantify the attack consequences. To construct the optimal defense against an FDIA, this paper examines the characteristics and construction of FDIAs from the perspective of attackers, aiming to reveal the shortcomings of traditional bad data identification. From the perspective of those defending against an FDIA, a two-layer defense model is proposed that includes detection and protection. First, information from multiple sources is utilized to improve the detection of false data. Second, extra PMUs are added as a protection method to strengthen the measurement redundancy. A zero-sum static game algorithm is applied to optimize the deployment of defense resources. The effectiveness of the proposed strategy is tested using IEEE 14-bus, 57-bus and 118-bus systems, and the impact of information asymmetry is also discussed.

KW - False data injection attack

KW - Game theory

KW - Optimal load shedding

KW - State estimation

U2 - 10.1016/j.ijepes.2018.07.007

DO - 10.1016/j.ijepes.2018.07.007

M3 - Journal article

VL - 104

SP - 169

EP - 177

JO - International Journal of Electrical Power & Energy Systems

JF - International Journal of Electrical Power & Energy Systems

SN - 0142-0615

ER -