Assessing the Impact of Network Topology on Seismic Resilience in Gas Pipeline Networks

Gas pipeline networks (GPNs) must be resilient to seismic hazards to reduce the risk of pipeline rupture after seismic events. Often modeled as complex graphs, these networks can exhibit varying topological structures. While network topology is assumed to impact seismic resilience significantly, this assumption remains unverified. This study investigates how network topology influences the seismic resilience of GPNs by comparing two common configurations: tree-like (Gaslib-134) structure and loop-like (Gaslib-135) structure. A total of 50,000 synthetic networks were generated for each type of network configuration based on different rehabilitation states. The resilience of GPNs was evaluated using three key metrics: robustness, reliability, and vulnerability, alongside nine topological attributes, including link density, node degree, central point dominance, articulation point density, modularity, heterogeneity, path length, network diameter, and spectral gap. Three statistical analyses, including correlation analysis between resilience metrics and topological attributes, box-plot comparisons of resilience metrics between network configurations, and an investigation into the variations in correlations under different damage scenarios, reveal significant relationships between seismic resilience and network topology. Results indicate that loop-like topologies generally exhibit greater serviceability, integrity, and robustness. In contrast, the ability of a tree-like network to achieve a higher level of robustness and connectivity is intrinsically constrained. These findings provide valuable insights for utilities to understand key features of GPN topology, guiding network expansion, and precise hydraulic analysis in future stages.