Optimization of Sewer Networks Using an Adaptive Genetic Algorithm

• Published in: Water resources management Vol. 26; no. 12; pp. 3441 - 3456
• Main Authors: Haghighi, Ali, Bakhshipour, Amin E.
• Format: Journal Article
• Language: English
• Published: Dordrecht Springer Netherlands 01.09.2012; Springer; Springer Nature B.V
• Subjects: Adaptive algorithms; Atmospheric Sciences; Chromosomes; Civil Engineering; Construction costs; Decision making; Design; Design engineering; Earth and Environmental Science; Earth Sciences; Earth, ocean, space; Environment; Exact sciences and technology; Flow velocity; Genetic algorithms; Geotechnical Engineering & Applied Earth Sciences; Hydrogeology; Hydrology. Hydrogeology; Hydrology/Water Resources; Linear programming; Materials handling; Mathematical programming; Methods; Metric system; Networks; Optimization; Sewer systems; Sewers; Studies; Water resources; Water resources management; Constraint handling; Adaptive genetic algorithm; Optimization; Sewer networks; algorithms; models; cost; reliability; Sewers; slopes; indicators; networks; hydraulics; channels; optimization; water resource management; strategy; methodology
• ISSN: 0920-4741; 1573-1650
• DOI: 10.1007/s11269-012-0084-3

This work aims at introducing an optimization model to design sewer networks. The approach specially focuses on handling the nonlinear and discrete constraints of the problem. For this purpose, an adaptive genetic algorithm is developed so that every chromosome, consisting of sewer diameters and slopes and pump indicators, is a feasible design. The binary chromosomes are freely generated and then decoded to feasible design alternatives following a sequential design-analysis algorithm. The adaptive decoding strategy is set up based on the open channel hydraulics and sewer design criteria. Through the proposed method, all the sewer system’s constraints are systematically satisfied. Consequently, there is neither need to discard or repair infeasible chromosomes nor to apply penalty factors to the cost function. A benchmark sewer network from the literature is considered to be designed using the proposed approach. The obtained results are then discussed and compared with the previous works. It is found that the adaptive constraint handling method computationally makes the optimization more efficient in terms of speed and reliability.