A Novel Method for Topology Preserving Static Network Reduction

Modern power grids become increasingly interconnected, and renewable energy sources amplify regional differences in energy generation and consumption. As a consequence, power grid analyses need to consider larger grids to examine the interaction between different regions. A detailed simulation of complete systems is often not feasible due to computational limitations. Thus, network reduction techniques are employed to create computationally less expensive equivalents of less important parts of the network. In this paper, we present a novel static network reduction method that is focused on the preservation of the overall network topology. We implemented this method in Python with DIgSILENT PowerFactory for the network modeling. We evaluate our method on a model of the German transmission grid and compare it to traditional reduction methods. In contrast to these traditional methods, our method yields interpretable equivalents with a preserved network topology consisting of standard components, while achieving a comparable reduction rate and accuracy of power flow and short circuit results.