Clustering of power networks: An information-theoretic perspective

An electrical power transmission system can be represented as a network with nodes and links representing buses and impedance between the buses, respectively. Decomposing a large interconnected power network into smaller loosely-coupled groups facilitates easy and flexible management of the power transmission systems by allowing secondary voltage control at regional levels and controlled islanding that aims to prevent the spreading of large-area blackouts. In this paper, we address the problem of clustering a power network into prespecified number of zones that are weakly coupled electrically by using an information-theoretic clustering algorithm. A notion of electrical similarity between electrical buses is developed, where any two buses are considered similar, if reactive perturbations at these buses have a similar effect on the entire network. A thorough theoretical justification of the use of the proposed clustering approach is provided along with the results of our methodology for IEEE test systems.