Topological Models Underestimate the Impact of Hazards on Coupled Power and Water Distribution Systems

As the number and severity of low-probability, high-impact weather events increases due to climate change, accurate models are needed to better understand the impact of hazards on critical infrastructure. Many critical systems, such as water and telecommunications, are dependent on the power network. At the distribution level, water and power systems are coupled in that water pumps are driven by three-phase induction motors. These motors require certain levels of power quality; however, hazard simulation literature does not accurately model power flow in distribution systems, opting for topological models to decrease computational burden and data collection. In this paper, we explore several case studies on a test system to show how including power flow both changes and adds to our understanding of distribution system vulnerability. We find that topological models underestimate pump failure in over 50% of single line outage cases. In addition, power flow models can be used to show the impact of other hazard events and failures, such as high impedance faults, voltage regulation failure, and secondary outages. On our test system, almost 50% of single high impedance fault cases result in pump failures. Our results show that topological models can underestimate pump failures, and therefore system vulnerability, in coupled power and water distribution systems.