Development of network simplification techniques for water hammer modeling

Abstract

Behaviour of water distribution systems under transient conditions is complex, making them inherently difficult to model. By developing ways of simplifying a pipe network model, the time taken to create an efficient and accurate transient model can be reduced. Interviews conducted with water hammer modellers from around the world, both academics and consultants, identified a number of simplification techniques currently employed. Such techniques are often applied without appropriate justification, with their effectiveness assessed solely on maximum and minimum pressures. Using the transient analysis program, TransAM, a number of case studies were conducted to determine the validity and effectiveness of such simplifications. Experimental tests were performed on the single pipe network in the Robin Hydraulics Laboratory at The University of Adelaide, to investigate the effects of truncating sub-networks. These tests validated the TransAM model developed and strengthened the conclusion that it is better to truncate sub-networks partially rather than back to the main pipeline. Effects of network skeletonisation were investigated on the Willunga water distribution network, identifying that it is possible to reduce the number of pipes in the system by half without significantly affecting the global maximum and minimum pressures in the system. An investigation of the Clare Valley network identified that tanks and sometimes pumping stations act to segregate networks, allowing them to be split-up into a number of smaller sub-systems.