Article ID: | iaor20105479 |
Volume: | 41 |
Issue: | 4 |
Start Page Number: | 297 |
End Page Number: | 311 |
Publication Date: | Apr 2009 |
Journal: | Engineering Optimization |
Authors: | Geem Zong Woo |
Keywords: | networks: flow |
The optimal design of water distribution networks is a non-linear, multi-modal, and constrained problem classified as an NP-hard combinatorial problem. Because of the drawbacks of calculus-based algorithms, the problem has been tackled by assorted stochastic algorithms, such as the genetic algorithm, simulated annealing, tabu search, shuffled frog-leaping algorithm, ant colony optimization algorithm, harmony search, cross entropy, and scatter search. This study proposes a modified harmony search algorithm incorporating particle swarm concept. This algorithm was applied to the design of four bench-mark networks (two-loop, Hanoi, Balerma, and New York City networks), with good results.