Layout and Size Optimization of Sewer Networks by hybridizing the GHCA Model with Heuristic Algorithms

Authors

1 School of Civil Engineering, Iran University of Science and Technology, P.O. Box: 16765-163, Narmak, Tehran, Iran

2 School of Civil Engineering & Enviro-Hydroinformatic COE, Iran University of Science and Technology, P.O. Box: 16765-163, Narmak, Tehran, Iran

3 Department of Civil Engineering, Faculty of Engineering, Isfahan University, Postal Cod: 81746-73441, Isfahan, Iran

Abstract

In this paper, a General Hybrid Cellular Automata (GHCA) model is hybridized with two of the most reliable heuristic search methods, namely Genetic Algorithm (GA) and Ant Colony Optimization Algorithm (ACOA), for the simultaneous optimal design of layout and size of pumped and/or gravity sewer networks. GHCA model is recently proposed by the authors for the optimal size determination of the sewer network with fixed layout. The model has shown to be able to optimally design pumped and/or gravity sewer networks, if required. In proposed hybrid models, the heuristic search algorithms are used to create trial layout for the network while GHCA is used to design the network by determining the pipe diameters, pipe slopes, drop height and pump height, if required. An ad-hoc engineering based method is used to determine feasible layouts by GA, while a Tree Growing Algorithm (TGA) is used to construct feasible layout using ACOA. The proposed hybrid models are tested against two benchmark sewer networks and the comparison of results to those of some existing methods indicates that proposed models, and in particular the ACOA-GHCA method, are more efficient and effective than some alternative methods for the optimal design of layout and size of sewer networks.

Keywords

Scientia Iranica
Volume 22, Issue 6 - Serial Number 6
Transactions on Civil Engineering (A)
November and December 2015
Pages 2069-2069

Files

History

  • Receive Date: 28 February 2015
  • Revise Date:
  • Accept Date: 09 July 2017

Share

How to cite

Statistics