Abstract
Water supply systems (WSS) are lifelines of communities as they enable security, health, and economic prosperity. Pipeline infrastructure in many older regions of the U.S. has deteriorated leading to significant leakages and frequent component failures. These issues currently threaten the supply reliability goals of water utilities and it is important to take cognizance of these vulnerabilities and develop appropriate response strategies. In this paper, a framework for optimized near real-time scheduling for operation and control of WSSs is proposed and demonstrated. The operational statuses of different types of valves and pumps that are inclusive of the system will be controlled based on an evolutionary optimization algorithm that is driven near real-time system monitoring data (e.g. tank level values). Energy efficiency and leakage minimization goals may also be accomplished through better control of system operations using monitoring data. The results of the optimization algorithm will lead to improved system operations and sustainable usage of critical resources such as energy and treated water. The proposed approach is demonstrated using a modified version of Anytown WSS. A genetic algorithm optimization code written in MATLAB is integrated with EPANET programming toolkit to enable the modeling of monitoring-data-driven optimal WSS control. This study and its findings would help water utilities in controlling the operations, maintaining reliability, and planning rehabilitation work of their systems.