Utilização de um algoritmo genético híbrido na operação de sistemas de abastecimento de água com ênfase na eficiência energética

• Ano de defesa: 2010
• Autor(a) principal: Costa, Luis Henrique Magalhães
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Palavras-chave em Português: Recursos hídricos; Algoritmos genéticos; Abastecimento de água
• Link de acesso: http://www.repositorio.ufc.br/handle/riufc/17092

Use of hybrid genetic algorithm in the operation in water supply system considering energy efficiency. Fortaleza, 2010. 146 p. Thesis (Doctorate) - Federal University of Ceará, Fortaleza, 2010. In general, operational rules applied to water distribution systems are created to assure continuity of the public water supply, without taking into account variations of the energy costs during a day. This causes an elevation of the energy costs due to the pumps. Furthermore besides rational use of energy by the pumps, there are other aspects which should be considered in order to achieve an optimized operation of a water transmission system, such as the daily variation of the water demand and the requirements regarded minimum and maximum water levels in the tanks and pressure requirements in the nodes of the water network. The objective of the present work is to develop a computer code which will determine on optimized operation rule for the system which will reach minimum costs of energy used by the pumps. The system is based in the use of Genetic Algorithms (GA) and the hydraulic network computer system EPANET. The GA for of the system is responsible for the search for rules of low energy costs and the hydraulic calculations are done by EPANET. Besides, one major innovation proposed by this research is the introduction of the Hybrid Genetic Algorithm which in order to reduce the stochastic standard aspect of the GA. The proposed methodology was applied to three study cases: two hypothetical and one real which was located in the city of the Ourém, Portugal. The results of these three study cases clearly show the superiority of the hydrid GA over the standard GA. The hybrid GA not only obtained better solution but also took much less time to run. Finally, it is expected that the use of this methodology will lead to more real time applications.