Sensitivity Analysis for Performance Evaluation of a Real Water Distribution System by a Pressure Driven Analysis Approach and Artificial Intelligence Method

Proper performance of water distribution networks (WDNs) plays a vital role in customer satisfaction. The aim this study is to conduct sensitivity analysis evaluate the behavior WDNs analyzed by pressure-driven (PDA) approach and classification technique using an appropriate artificial neural network, namely Group Method Data Handling (GMDH). For purpose, divided into four distinct steps. In first second steps, real network has been Pressure-Driven Analysis obtain pressure, ? coefficient, percentage supplied flow. performed three different values design peak coefficient k*. third step, (GMDH) applied several binary models have constructed. carried out input data, including topology base demand necessary satisfy requests users average conditions assuming that each single one-hour time step depends on coefficient. Finally, results obtained from PDA hydraulic those them GMDH algorithm compared out. innovation demonstrate parameters adopted are correct. confirms gives proper for case stable also when value k*, characteristic hour varies admissible technical range. It was confirmed approach, GMDH-type can provide higher sufficiency evaluation WDNs.