A Graph Theory-Based Clustering Method for Power System Networks

With the increasing human population and requirements, the current electric power grid is becoming complex and expanding rapidly. Increasing integration of renewable energy sources and other advanced metering technologies introduces a complex, expanded and distributed power system infrastructure. Real-time operation and control of large-scale power system networks have become a challenging task due to these reasons. Power system clustering is a feasible solution to overcome these challenges. Clustering utilizes parallel processing capabilities of control methods and provides adequate methods for real-time power system control. In this paper, a graph theory-based clustering approach is presented for large interconnected power system networks. The algorithm is tested on multiple power system networks from small to large scale and the performance is analyzed. Typical results indicate that the method can provide clusters with uniform sizes within a short time limit for any type of power system network. The clusters observed are applied in solving optimal power flow (OPF) problem.