Complex systems techniques applied to transmission expansion planning. Part I: generating random networks that are consistent with power transmission

Abstract

Transmission Expansion Planning (TEP) could benefit from studying how network structure affects power grid performance. To accomplish this, we require large ensembles of power grid data. These are not available, so it is therefore necessary to generate random networks that have a structure that is consistent with power grids. We propose a generative model that is able to capture the main characteristics of the power grid. This model is based on an epsilon-disk algorithm that incorporates the spatial component of power systems while taking into account the particular characteristics of sources (power plants), sinks (demand) and edges (transmission lines) in the problem. We study the performance of the algorithm and apply it to a real case study based on the Spanish power system.