CPS: Small: Data-driven Real-time Data Authentication in Wide-Area Energy Infrastructure Sensor Networks

University Of Tennessee Knoxville, Knoxville TN

Investigators

Abstract

The electric power grid is a large-scale dynamic cyber-physical systems (CPS) and it is critical to the modern society. To ensure its reliable operation, wide-area monitoring and control systems are essential to provide real-time understanding of the system dynamics such that proper control actions can be taken in real time to respond to power system disturbances and avoid blackouts. Like many other CPSs, power grids are vulnerable to data spoofing attacks due to the lack of security properties in monitoring and control protocols. Conventional cryptography techniques are not usable because the computation burden and time delay are intolerable to numerous power system devices considering the fast-changing system dynamics. This project proposes to develop spatial and temporal signature-based methods and tools that efficiently authenticate power grid wide-area measurements and effectively prevent malicious cyber-attacks from causing catastrophic losses. By delivering data authentication tools to detect cyber-attacks, this project will provide a new path that leverages measurement data from CPS itself to protect the security of wide-area highly-dynamic closed-loop real-time CPS. This project will have immediate positive impacts on electric power grid security and will also be beneficial to other tightly-coupled, physically-distributed CPSs. The project will also provide many educational resources and opportunities for K-12 and STEM students to broaden their participation in engineering. The project's primary goal is to address power system cyber security issues and help protect the power system from data spoofing attacks that are hard to detect by existing technologies. The project team discovered that power grid wide-area monitoring system (WAMS) measurement data have spatial and temporal signatures caused by natural responses of each local grid's continuous and random condition changes. These signatures are almost impossible to counterfeit, making them perfect for data authentication. This project will be the first time to use signatures in high data rate (up to 1,500 sample per second) measurements to develop data-driven cyber-physical security tools. Major steps to achieve the project goal include: (1) investigate methods to extract spatial and temporal signatures embedded in WAMS measurements efficiently; (2) develop methods and tools for authenticating a large volume of WAMS measurement data in real time; and (3) online implement and demonstrate cyber-attack detection technologies on FNET/GridEye, which is an actual cyber-physical system. With credible real-time information provided by the secure cyber system, power system operators will be able to assess system risks or take prompt actions to control and mitigate risks and prevent blackouts. The secured cyber system will also ensure the physical power system is under appropriate controls determined by automatic controllers to prevent systems entering dangerous status. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.