Collaborative Research: Enabling Safe and Efficient Restoration of Drinking Water Systems through Multiphase Flow Modeling and Control

Drinking water service is expected to be restored quickly after water mains are repaired, new lines are installed, or communities expand. Yet refilling an empty water main can create hidden hazards. As water enters the pipe, trapped air is compressed and displaced, producing damaging pressure surges that can cause additional pipe damage, water loss, longer service interruptions, and increased repair costs. Filling the system more slowly reduces these risks but delays the restoration of water service to homes, businesses, hospitals, and other essential facilities. The research funded by this award will develop new scientific understanding and practical engineering tools to help water utilities restore service more safely, reliably, and efficiently following repairs and construction. The project will also support education and workforce development by integrating research into university courses, training students in water infrastructure and data-enabled engineering, while promoting technology transfer through collaborations with municipal utilities and laboratory testbeds. The project will develop computational methods for modeling, monitoring, and controlling multiphase air-water flow during the refilling of transmission mains and water distribution networks. The research will formulate computationally efficient models that capture the dominant physical mechanisms governing water motion, air compression, air release, and pressure dynamics without requiring computationally intensive multiphase flow simulations. These models will be calibrated using physics-informed neural differential equations that combine conservation-law models with sensor data. The project will further develop uncertainty-aware state estimation methods to infer unmeasured pressures, flows, air volumes, and filling states from limited sensor measurements, and will integrate these estimates into optimization-based control algorithms for pumps, valves, hydrants, and air-release devices. The methods will be validated through laboratory experiments, realistic network testbeds, and field-informed case studies conducted in collaboration with water utilities in Texas and Alabama. The expected outcome is a general computational framework for modeling, monitoring, and regulating multiphase flow during water system priming. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.