OPTIMIZATION OF PIPELINE NETWORKS WITH A HYBRID MCST-CD NETWORKING MODEL

In this paper, we present a hybrid network model (HY-PIPENET) that use s a minimum cost spanning tree (MCST) network algorithm to determine t he optimum path and constrained derivative (CD) method, to select an o ptimum pipe diameter. The HY-PIPENET has been validated with the publi shed data of a six-node/seven-pipe networking system, and, moreover, t his system has been optimized with the MCST-CD method. As a result, it was determined that the gas can be sufficiently supplied at a lower p ressure with smaller diameters of pipe, compared to the original syste m in a six-node/seven-pipe network. Hence, the construction cost was r educed about 40% in the optimized system. The hybrid networking model has also been applied to a complicated domestic gas pipeline network i n a metropolitan area of Korea. In this simulation, a parametric study was performed to understand the role of each individual parameter suc h as the source pressure, flow rate, and pipe diameter on the optimize d network. From the results of these simulations, we have proposed tha t the optimized network is a tree-type structure with an optimum pipe diameter and source pressure, which was used in a Korean metropolitan area. This proposed system however, does not consider potential enviro nmental problems as well as safety concerns.