Augmented Gradient Method for Head Dependent Modelling of Water Distribution Networks

When analysing a pressure deficient network, it is crucial that the pressure dependent nature of nodal outflows be taken into account. The head dependent analysis (HDA) produces an accurate representation of the nodal outflows and network hydraulic performance. This is essential when modelling pipe leakages, network redundancy and reliability. These are vital aspects often considered in the optimization of a water distribution system (WDS). This paper describes an approach for head dependent analysis in which an embedded function for the head-outflow relationship is incorporated in the Gradient Method (GM). The procedure is capable of simulating both normal and deficient network operating conditions effectively. Results based on networks from the literature show that the proposed method is robust and converges smoothly and rapidly.