Feasibility of Mass Balance Approach to Water Distribution Network Model Calibration

The increase of real water losses in water distribution networks (WDNs) is worrisome for social community and water utilities. Nowadays, enhanced WDN hydraulic models allow simulating pressure-dependent background leakages, thus, being of direct relevance to support water loss reduction actions. Unfortunately, traditional calibration approaches rely on demand-driven analyses and do not account for additional parameters of the pressure-dependent background leakage models. This work proposes a novel framework for calibrating enhanced WDN models which permits to pursue mass-balance (matching of flow observations at water sources) by accounting for pressure-dependent water demand components and demand patterns. At the same time, energy balance (matching of pressure observations) allows to get the realistic prediction of all water demands and background leakages. The novel calibration strategy involves the simultaneous estimation of pipe hydraulic resistances, parameters of the background leakage model and relevant customer demand patterns. The feasibility of the strategy is demonstrated on a real WDN located in Southern Italy. © 2017 The Authors.