Tradeoff Between Robustness and Functionality in Cyber-Coupled Power Systems

Robustness and functionality are conflicting requirements in cyber-coupled power systems. In general, a more tightly coupled power and cyber networks provides better functionality, but may degrade the robustness against attacks or failures. The way in which the power network is coupled with the cyber network is thus crucial to the design of robust cyber-coupled power networks while maintaining sufficient functionality. In this article, we classify coupling patterns according to two node-criticality metrics of the cyber network, i.e., node degree and node betweenness, and two node-criticality metrics of the power network, i.e., node degree and node capability. We use the relative coupling correlation coefficient to quantify the coupling pattern, and show that a coupled system with a lower relative coupling correlation coefficient has better robustness. A multiobjective problem is formulated and the Pareto optimal solutions are found to generate coupling patterns that give balanced robustness and functionality. Using a realistic physical power flow model, our results reveal possible tradeoff between functionality and robustness under different choices of criticality indexes.