EVOLUTIONARY COMPUTING TECHNIQUES FOR OPTIMAL PRESSURE REGULATION IN WATER DISTRIBUTION NETWORKS

Abstract

This paper addresses the problem of appropriate electrical motor valves setting for the pressure regulation of a water distribution networks for specified nodal demands by using both genetic algorithm (GA) and a relatively new concept known as Shuffled Complex Evolution-University of Arizona (SCEUA). To demonstrate the performance of both techniques, a simple illustrative example of a controlled water distribution networks is presented showing the effectiveness of both algorithms to regulate the pressure at all the network nodes, between upper and lower values and as near as possible to the target values. Regardless the final mathematical solutions of both algorithms which are approximately the same, results show the superiority of SCE-UA technique to reach the optimal solution using less number of function evaluations than GAs. This paper concludes that the SCE-UA algorithm is well suited to deal with water supply networks problems, which provides a rich field for future research.