| description abstract | Various water distribution leak detection and localization methods have been proposed for supervisory control and data acquisition (SCADA) data collection systems. However, because their available numbers of measurements are limited, the SCADA systems are often insufficient to identify realistic sized leaks. A clear next step is to develop detection and localization methods for smart systems that collect advanced metering infrastructure (AMI) data (i.e., AMI systems); however, only the authors have proposed tools for the AMI data collection systems. To encourage the usage of AMI data, this study tested five types of data collection systems for leak detection and localization that measure (1) only source flows, (2) source flows and a few nodal pressures, (3) source flows and AMI demands, (4) AMI demands and a few nodal pressures, and (5) AMI pressures and AMI demands. An appropriate leak detection and localization tool for each data collection system is applied and tested for two water distribution networks: one located in Austin, TX, and the other in Tucson, AZ. Each system’s performance was evaluated using metrics of detection probability, false alarm rate, time to detect, and localization pipe distance. Overall, based on the obtained results, the SCADA systems were poor in detecting realistic-sized leaks, while the AMI systems successfully identified those small failures. Thus, the AMI systems were required to improve detection, and a pressure-supplemented AMI system was necessary to obtain high localization performance, particularly for a large network, such as Austin, Texas. | |
