Attack Surface Modeling for Situational Awareness and Vulnerability Analysis of Power Systems

Complex networked systems, like the power grid, use feedback architectures as a means to change the native dynamics of system components and compensate for variation and uncertainty in the actual integrated behavior of these components. Nevertheless, these feedback cycles of information flow through the system create intrinsic vulnerabilities that can be exploited through stealthy, coordinated attacks on the system, or they may even be triggered by component failures or other disruptions. The result canbe catastrophic, with cascading failures ripping through the system, destroying components and shutting down operations. The aim of this project is to develop tools that can be used to understand and quantify the exposure of power systems to such intrinsic vulnerabilities. In Phase I, general vulnerability analysis tools were adapted to and shown to be effective on approximate, linear models of power systems. Attacks designed using these techniques appear to be catastrophic, stealthy, and mathematically generated: made-to-order at varying levels of attacker sophistication or ability to access the power system (e.g.,insider threatsvs. external attacks). A sophisticated mathematical technique was also developed for simplifying nonlin- ear models of power systems, with the potential for extending the linear tools to more realistic, nonlinear models. Phase II will further develop these tools by developing mitigation strategies based on linear mod- els, developing the computational methods to design made-to-order attacks based on nonlinear models, and applying machine learning methods to attack design. The tools developed in this project will allow system operators to identify the disruptions, including attacks, to which the system is most vulnerable, quantify the associated risk, and develop mitigation strategies. A systematic risk analysis service will simultaneously reduce operating margins while guarding against component failures and external attacks. Protecting the power grid in this way has the potential to save billions of dollars and hundreds of human lives, as evidenced by the 2021 Texas Power Crisis.