Pressure determination in water distribution systems (WDS) is important because it generally drives the operational actions for leakage and failure management, backwater intrusion and demand control. This determination would ideally be done through pressure monitoring at every junction in the distribution system. However, due to limited resources, it is only possible to monitor at a limited number of nodes. To this end, this work explores the use of an Artificial Neural Network (ANN) to estimate pressure distributions in a WDS using the available data at the monitoring nodes as inputs. The optimal subset of monitoring nodes are chosen through an entropy-based method. Finally, pressure values are compared to synthetic pressure measures estimated through a hydraulic model.
Artificial Neural Networks and Entropy-based Methods to Determine Pressure Distribution in Water Distribution Systems
Ridolfi, Elena;
2014
Abstract
Pressure determination in water distribution systems (WDS) is important because it generally drives the operational actions for leakage and failure management, backwater intrusion and demand control. This determination would ideally be done through pressure monitoring at every junction in the distribution system. However, due to limited resources, it is only possible to monitor at a limited number of nodes. To this end, this work explores the use of an Artificial Neural Network (ANN) to estimate pressure distributions in a WDS using the available data at the monitoring nodes as inputs. The optimal subset of monitoring nodes are chosen through an entropy-based method. Finally, pressure values are compared to synthetic pressure measures estimated through a hydraulic model.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
