Publication Type

Journal Article

Version

acceptedVersion

Publication Date

9-2018

Abstract

Load frequency control (LFC) is widely employed to regulate power plants in modern power generation systems of smart grids. This paper presents a simple and yet powerful type of attacks, referred to as resonance attacks, on LFC power generation systems. Specifically, in a resonance attack, an adversary craftily modifies the input of a power plant according to a resonance source (e.g., rate of change of frequency) to produce a feedback on LFC power generation system, such that the state of the power plant quickly becomes instable. Extensive computer simulations on popular LFC power generation system models which consist of linear, non-linear, and/or high-order items clearly demonstrate the effectiveness of the proposed attacks. As the attack has very low computational cost and communication overhead, it is easy to launch in resource-limited devices such as intelligent electronic devices. In our simulations, the attacker keeps modified input within the normal operating range so as to invade plausibility and consistency based attack detection methods and yet the modifications can quickly drive the system beyond the admissible boundary. Another interesting finding is that by maliciously modifying the input such as power load and tie-line signal over multi-area interconnection channels, a multi-area LFC power generation system could become unreliable more quickly than a single-area system. Finally, we propose countermeasures on the proposed attacks.

Keywords

Load frequency control (LFC), rate of change of frequency (RoCoF), cyber-physical system (CPS) security, false data injection (FDI), system stability

Discipline

Information Security

Research Areas

Cybersecurity

Publication

IEEE Transactions on Smart Grid

Volume

9

Issue

5

First Page

4490

Last Page

4502

ISSN

1949-3053

Identifier

10.1109/TSG.2017.2661307

Publisher

Institute of Electrical and Electronics Engineers (IEEE)

Copyright Owner and License

Authors

Additional URL

https://doi.org/10.1109/TSG.2017.2661307

Download

Share

COinS