Services on Demand
Journal
Article
Indicators
Cited by SciELO
Access statistics
Related links
Similars in SciELO
Share
Tecnología y ciencias del agua
On-line version ISSN 2007-2422
Abstract
LERMA, Néstor; PAREDES-ARQUIOLA, Javier; ANDREU, Joaquín and SOLERA, Abel. Comparison of Algorithms for the Optimization of Water Resources Systems. Tecnol. cienc. agua [online]. 2015, vol.6, n.3, pp.79-92. ISSN 2007-2422.
Techniques to obtain operating rules (OR) for multi-reservoir systems have consisted of using simulation models to evaluate the behavior of the system after the OR has been defined. This is a long and repetitive process lasting until the appropriate OR is found. The objective of the article herein is to compare two evolutionary algorithms used to obtain the OR. To this end, a methodology has been developed which combines evolution algorithms with flow networks. The latter are used to simulate the management of a system based on the definition of some of the OR, and evolutionary algorithms modify these rules and control successive simulations. The SCE-UA and Scatter Search algorithms were compared and the SIMGES flow network module was used, which is part of the AQUATOOL Decision Support System. The methodology proposed was applied to the Tirso-Flumendosa-Campidano system on the Island of Cerdeña, Italy. This system is located in a Mediterranean climate characterized by changes in the availability and demand of the resource. This type of system is considered to be among the most complex to manage. The results obtained show that optimal OR can be obtained with the proposed methodology, and thus using the evolutionary algorithms (one more appropriate than the others). The OR obtained by this study resulted in lower deficit-related demands and pumping costs than with the current management.
Keywords : Evolutionary algorithms; Water System; AQUATOOL; SIMGES; Management; Optimization; Mediterranean; Operating Rules.