Abstract

Abstract—It is well known that information will play an impor-tant role in enhancing emerging smart grid system operation. Therefore, questions naturally arise as to when the increased data-dependence may be considered excessive. Two practical considerations emerge: 1) communications and computational overhead, in which redundant and irrelevant information acqui-sition and use results in heavy computational burden with limited performance return; and 2) increasing risks of power system dis-ruption due to information delay from communication congestion or cyber attack. One strategy to improve smart grid resilience is to determine the appropriate degree of dependence on cyber information to balance performance with overhead and risk. In this paper, we present a hierarchical cyber-physical multiagent model of smart grid system operation based on flocking theory in the context of the transient stability problem. Through this model, we study strategies that harness a selective degree of cyber tech-nology by leveraging physical couplings. Our formulation enables the identification of large-scale distributed control strategies for robust and resilient power grid operation. We demonstrate the potential performance improvements of our findings on the New England 39-bus power system for case studies involving a variety of system faults and communication delays. Index Terms—Cyber-physical modeling of smart grid, flocking-based dynamical systems, power system transient stability. I.