
Experimental and Numerical Investigation of Pressure Transients Induced by Water Consumption Variations in Complex Water Distribution Networks
In pipe networks, pressure plays a fundamental role: pressure variations induced by transients can generate additional stresses not only on pipes, but also on junctions and hydraulic devices. In Water Distribution Networks (WDNs), however, the effects of transients are often underestimated, based on the assumption that these systems are intrinsically self-protect. A large body of literature addresses the transient behavior of WDNs through numerical analyses. Conversely, experimental investigations based on laboratory and field tests remain scarce. It is necessary to derive general conclusions on the role of the transient source location, as well as on the magnitude and simultaneity of consumption variations, for a given network layout. To fill this gap, the present research focuses on laboratory experiments aimed at isolating and analyzing transients induced by changes in users’ water consumption—the most frequent source of pressure variations in WDN.
Analyze the dynamic response of real systems to transients induced by a change in water consumption.
Develop a numerical model, based on the Method of Characteristics, as well as the Admittance Matrix Method, to reproduce the dynamic behavior of complex systems in transient regime, with particular interest in plastic materials.
Analyze the effect of a real leak in a plastic pipe in unsteady-state conditions.
Further investigate transient test-based techniques for leak survey in plastic pipelines, in order to develop a machine learning model for locating leaks in a looped water distribution network using minimal number of leak scenarios.
Conduct a literature review on transient analysis in complex systems.
To analyze the dynamic response of real systems to transients caused by a change in water consumption, a wide experimental program will be carried out in the Water Engineering Laboratory (WEL) of the University of Perugia. The numerical model that will be employed to model the experimental water network and examine the water hammer phenomena is the Method of Characteristics and the Admittance Matrix Method. Specifically, the Admittance Matrix Method exploits graph theoretic concepts to organize the 1-D water hammer governing equations, perturbed, linearized and solved in the frequency domain, in a matrix form that deals only with nodal variables.
To obtain a better understanding of the behavior of water networks in transient conditions
To develop a numerical model that reliably simulates losses and transient phenomena in a complex network
To give the first experimental validation of the admittance matrix method on a complex system in order to assess its accuracy in modeling transients and perform calibration for the diagnosis of the system.