Multi-objective optimization of water distribution system: a hybrid evolutionary algorithm
Masoud Gheitasi, Hesam Seyed Kaboli, Alireza Keramat
Abstract
The optimal operation of water distribution systems is complicated due to multiple objectives that are in conflict, such as water quality versus cost. This work proposes to combine Strength Pareto Evolutionary Algorithm (SPEAII) with Multi-Objective Particle Swarm Optimization (MOPSO) algorithm, called MOPSO-SPEAII, to establish a multi-objective model which handles water quality, costs and storage-reliable requirement. The main idea is that genetic operators are combined with particle swarm operators such that the fitness of SPEAII results is evaluated using MOPSO. An optimization-simulation model is prepared by linking the hybrid algorithm with EPANET software, and it is employed for a typical case study from the literature. The model outcomes verify that the MOPSO-SPEAII is more stable compared to SPEAII in terms of closeness to global minimum and can be used as a robust decision tool. However, the model application for a real sized system increases the computational intensity of the model.