A Hybrid Algorithm Based Approach for Optimization of Real Time Piped Flow Water Distribution Network to Minimize the Cost of Project Implementation

Abstract

Water Distribution Systems (WDS) are one of the major infrastructure assets of the society, it is important to design economically effective water distribution system for minimizing operating cost and ease of maintenance. Designing economically effective WDSs is a complex task, which involves optimizing sizes, locations, and operational statuses of network components such as pipes, pumps, tanks and valves. The objective of this study is to minimize the implementation cost of real time water distribution network (WDN)by minimizing the diameters of the pipes. In this study first EPANET 2.0 analysis of the target WDN is carried out to determine the pipes size. The diameters of the pipe required for EPANET analysis is obtained by theoretical calculations.  The system is analyzed using EPANET 2.0 hydraulic solver to determine the pressure heads at each node and velocities in each pipe. Hybrid algorithm which is combination of GA and PSO is developed and implemented in MATLAB to analyze the target WDN. The required pressure head at each junction and velocities obtained from the EPANET Hydraulic solver is given to Hybrid algorithm for the further analysis. The developed algorithm will use combination of GA and PSO optimization technic to select optimal pipe sizes by taking care of minimum required pressure heads and other operating conditions at each nodes. The resulted obtained by this analysis is compared with that of previous study which is of pure GA approach. Based on the analysis WDN is redesigned using lower level pipe sizes and hence minimizing the cost of pipes. Cost of pipe plays major role in minimizing the implementing cost of water distribution network projects and it is observed that there is a significant around 2.4% reduction in pipe cost and around 1% reduction in project cost.