PipeNetGen: A method for generating synthetic pipe networks with open source data

Abstract Submission: Water distribution networks are centralized systems responsible for delivering water across vast geographic areas to diverse user groups. Given their critical role in society, significant emphasis is placed on optimizing their efficiency and enhancing water conservation. However, security concerns prevent water utilities from publicly releasing their network models, limiting opportunities for in-depth study. Additionally, smaller water utilities may not have the resources to produce a hydraulic model. These challenges limit researchers' ability to model specific city water distribution systems and simulate new water management strategies. Previous research has developed methods for generating synthetic water distribution networks using open-source data, graph theory, and Mixed-Integer Linear Programming (MILP). Although these methods have produced feasible networks, they have not yet been validated against actual water utility hydraulic models. This research introduces the Pipe Network Generator tool (PipeNetGen), which can create a synthetic pipe network for any city using open-source data. PipeNetGen includes a core MILP model based on previous research and combines it with a clustering algorithm to run larger systems, a hydraulic simulator to correct MILP assumptions, and a fire flow process to ensure network reliability under high demands. PipeNetGen is applied to Lakewood, California, where the resultant synthetic network is compared with a water utility-provided network regarding network redundancy, hydraulics, and demand. This tool is also applied to cities of varying sizes to demonstrate its high versatility. This tool provides a valuable resource for researchers requiring a water infrastructure model for their city of study.

Learning Objectives/Expected Outcome (Optional) :