Author(s): Zheng Y. Wu
Linked Author(s):
Keywords: Water distribution systems; Genetic algorithms; Optimal design; Optimal rehabilitation
Abstract: This paper presents an approach for discrete optimal design and rehabilitation of water distribution systems. A simple analysis showing a comparison of continuous and discrete optimization is presented. It indicates that there is no guarantee, in general, that the continuous optimization model using a fitted cost function will reach the least cost design solution. A discrete pipeline optimization model has been proposed and implemented by integrating a genetic algorithm with a hydraulic network simulator. The model is formulated to search for the least cost combination of available pipe sizes. It has been tested on an example for the optimal design of water distribution systems. The results have shown that the discrete optimization model developed in this paper is effective at producing the optimal and near optimal solutions.
Year: 2001