After leaving water treatment facilities, drinking water is delivered to consumers through an intricate network of pipes. Recent studies have demonstrated that the quality of treated drinking water deteriorates during the journey from treatment plants to consumers? taps; this has significant implications for public health. Operating under limited budgets, water utilities adopt different degrees of water quality monitoring that are geared towards regulatory compliance but are unbefitting for rapid detection and mitigation of contamination events. Furthermore, utilities typically lack targeted, real-time control strategies and often respond to contamination events by issuing utility-wide advisories (e.g., ?boil water? or ?do not consume? advisories). Such advisories have significant socio-economic impacts and are typically slow in improving water quality. In contrast to this, this project creates new methods for real-time water quality management in urban water networks by controlling hydraulic pumps, valves, and disinfectant dosing stations in response to contamination events. To this end, this research harnesses recent advances in water sensing technologies while investigating control algorithms through interdisciplinary research from environmental science, optimization, network control, and aquatic chemistry. This research enables real-time monitoring and control of hydraulics and water quality, allows water utilities to adopt strategies in response to contamination events, and broadens participation of underrepresented groups in research and education at UT San Antonio, UT Austin and the University of Illinois at Chicago.
This project puts forth a novel mathematical framework that couples the hydraulic equations governing water flow and pressure with dynamic water quality models depicting the transport and decay of disinfectant residuals, which act as a proxy for contamination event detection in drinking water distribution networks. The resulting framework faithfully describes the network operation under regular conditions and contamination events. This framework also enables the development of scalable optimization algorithms that are amenable to real-time implementation for control of pumps, valves, and disinfectant booster stations, thereby ensuring compliance with water quality standards. The algorithms are designed to deal with a variety of water system applications such as flow modulation, response and recovery from contaminant intrusion events, and reliable network-wide disinfection. The theory is evaluated on realistic water network models in addition to data from fixed and mobile sensors that collect hydraulic and water quality data from a real-life water system.
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.
