Efficient Numerical Calibration of Water Delivery Network Using Short-Burst Hydrant Trials

Well-calibrated hydraulic models enable utilities to detect hidden leaks, evaluate emergency scenarios, and plan infrastructure upgrades with confidence. Calibration involves adjusting uncertain parameters. Although modern meters with Internet of Things (IoT) devices now capture demands and valve settings in detail, pipe roughness is still very difficult to measure directly. This study presents a practical method using short night-time hydrant trials—controlled high-flow discharges that temporarily increase velocities and head losses—to estimate pipe roughness in real-world WDNs. During each trial, a few portable pressure loggers collect high-frequency pressure head data, and the discharge rate is recorded at the hydrant. By conducting several such trials in strategically selected locations, increased flow and head losses are induced across subregions that uniformly cover the WDN zone. The resulting data set allows optimization algorithms to adjust roughness values more effectively. The entire workflow can be completed in one night of fieldwork and computation. Yet, it matches or surpasses the accuracy of extended daytime surveys, even in systems with oversized pipes and low-pressure differences. Requiring only a few dependable pressure loggers and smart meter readings, the approach is applicable for utilities of all sizes and promises faster, more reliable model calibration.