Note: This content is accessible to all versions of every browser. However, this browser does not seem to support current Web standards, preventing the display of our site's design details.


Cyber Security in Power Networks



Peyman Mohajerin Esfahani

    Computation, communication, sensing and actuation devices connected over computer networks can improve operation and performance in power systems, but this increasing reliance on cyber components makes the system vulnerable to attackers who could access these components through the computer network and thereby cause significant physical and economic damage. To prevent such scenarios, fault detection and isolation (FDI) methods for dynamical systems can be used. These methods generate a diagnostic signal based on measured signals and a mathematical model of the system that is sensitive to the occurrence of specific attacks or faults. One essentially designs a filter using all available information as inputs and creates a map from attack signals to a residual that is, ideally, non-zero when the system is under attack and zero otherwise. In practice, due to uncertainty and noise, the FDI filter will declare an attack only when the residual surpasses a predefined noise threshold. Previous research at IfA has investigated variations of this problem.

Project Context and Goals
    In this project, the student will take the role of a "white hat hacker" to assess the amount of damage that could be done to the system by an attacker while the system is monitored by an FDI filter. In particular, the goal is to design a strategy for the attacker to maximally disturb the system while the filter residual remains below the detection threshold. The associated damage and/or increased operating cost should be quantified as a function of various system and filter parameters. Game theoretic scenarios and formulations between power system operators and potential attackers can be considered and studied in depth for MA projects. The student will become familiar with recent FDI techniques both for continuous and discrete time dynamic models, power system modeling, and application of the FDI scheme for cyber security issues in power systems. Furthermore, the student should implement the techniques in simulation on a test network (e.g., the IEEE-118 bus network).

  • The project combines rigorous mathematical aspects with practical research and would be a good experience for those wishing to go to industry.


John Lygeros

Art der Arbeit:
Voraussetzungen: Signals and Systems I and II, MPC, and proficiency with MATLAB.
Anzahl StudentInnen:
Status: open