Optimal Robustness in Power Grids From a Network Science Perspective

In recent years, the frequent occurrences of large-scale blackouts highlight the necessity of enhancing the robustness of power grids. In this paper, based on the cascading failure model which considers both the network topology and electrical characteristics running in power grids, we study how the topological metrics affect the robustness of power grids. The topological metrics considered include both the connectivity of the network and the distribution of generators. We compare the results in typical complex network models and IEEE 118-bus network. Moreover, by using the simulated annealing method, we find the optimal network topology to achieve the best network robustness, and compare the topological metrics before and after the optimization to verify our findings. The results show that in order to enhance the robustness of power grids, it is better to make the network sparsely connected, and place the generators as hubs and decentralize these generators.