Optimizing leakage management in water distribution networks using pressure-driven demand modeling and digital twin technology: a case study of the University of Macao Hengqin Campus

This study presents a data-driven leakage management system for the University of Macao Hengqin Campus water distribution network. The proposed framework combines online pressure/flow monitoring, a digital twin analytics platform, a pressure-driven demand (PDD) hydraulic model, and a digital work order system to support closed-loop workflows from detection and dispatch to repair confirmation and performance verification. Multi-scenario pressure regulation and controlled leakage tests were performed to quantify pressure–leakage relationships and calibrate leakage parameters for the campus network. For undetected leakage in predominantly metallic mains, a pressure exponent of 0.9 was identified. The calibrated PDD model reproduced monitored hydraulic states (pressure and flow) with average errors below 5%. Scenario-based inverse simulations, constrained by measured data, narrowed the plausible location of a small leak (≈0.5 m3/h). Pressure management analyses further revealed the trade-off between leakage reduction and maintaining minimum service-pressure requirements at critical nodes. By integrating analytics, decision support, and digital work orders, the platform enables end-to-end traceability and faster operational response. In the campus deployment, the detection-to-validation cycle was completed within 2 weeks, demonstrating improved responsiveness and intervention accountability. The results indicate the practical value and scalability of combining PDD-based modeling with digital twin platforms for proactive leakage control.