Synthetic modeling of water distribution systems for interdependent infrastructure systems resilience analysis with interdependencies via building-mediated clustering

Resilience studies for water distribution systems (WDS) coupled with other interdependent infrastructure systems attract increasing attention from stakeholders and researchers. However, most existing large-scale WDS models are single infrastructure-based without consideration of other infrastructure systems. This is due to a lack of needed information on systems coupling, the structure of the simulator used, and the computation load involved. To address these gaps, this paper presents a synthetic modeling framework for a real-world WDS as coordinating with other infrastructure systems via a building-mediated clustering approach through consideration of physical distance and node capacity. First, the WDS network topology and operation parameters are inferred via bulk open-source information. A building-mediated clustering approach is designed to systematically derive the interdependence between the WDS and the power system similarly created as a case study. Second, a novel linearization method is developed in formulating the WDS model that can relieve computation load while maintaining accuracy. Finally, a disruption-recovery framework is developed to demonstrate the proposed methodology in modelling WDS resilience. The framework is applied to a neighborhood in Queenstown, Singapore, an area of 20.43 km2 and 96,000 population. The near-real-time simulations on the coupled system involving 308 nodes and 384 links showcase the effectiveness and application of the proposed synthetic modeling and formulation.