Structural characterisation for the synthesis of large-scale combined electric–gas networks

Abstract

In this paper, the authors present a methodology for building realistic synthetic test cases to model combined electric and natural gas infrastructure networks. Because these networks are synthetic, that is, fictitious, they are able to be freely shared; hence, they serve to support research and development in the area of coupled-infrastructure analysis with large-scale realistic test cases that do not contain non-public information. We anchor our process for building these synthetic networks in a structural characterisation of actual electric and natural gas networks. Supply and demand nodes are geographically placed, based on a combination of publicly available information from several sources and a clustering-based method to match the right fraction of each node type, taking into account the intersection points between the electric and gas grids. Then the networks are connected with pipelines and transmission lines using a systematic graph construction method, validated against network properties including degree distribution, clustering, graph diameter, and degree of planarity, along with operational validation to ensure the simulated solution is realistic. The methodology is demonstrated by creating and validating a test case with 6717 electric buses and 2451 gas nodes.