Abstract:
Particular emphasis is given to the interest in network simulation expedients (skeletonization, load consolidation, and assumed pipe resistance factors). An approach to fulfilling the research need is presented which consists of building detailed models of actual but representative grid systems, verifying the performance of the models, and using them as research tools to investigate the impact of various simplifying assumptions often made in network modeling. With increased emphasis on digital computer modeling of the hydraulic performance of water-distribution networks for designand automatic operational control, there has been a recent upsurge of interest in the impact of expedients commonly employed for network simulation, i.e., skeletonization, load consolidation, and C value allocation. This article offers a historical explanation of the recent upsurge of interest, presents one approach to studying the impact of simulation expedients, demonstrates the value of the approach with a case study, and discusses the merits of the approach in terms of ultimately being able to generalize about the degree of input data refinement necessary to model grid systems adequately. The basic technique for hydraulic network balancing based on Hardy-Cross theory adapted to the digital computer has been available in the US since 1957.1 The first applications of the technology were for design of extensions and reinforcements to large distribution systems serving hundreds of thousands of customers.2 These systems were the first ones studied for two reasons. First, they represented the most critical problems facing network designers to which the technology could be applied. Second, digital computers were most available in the population centers. It is important to note that many of the early systems studied and reported in the journals were characterized by looped
