Meu SciELO
Serviços Personalizados
Artigo
Indicadores
Citado por SciELO
Links relacionados
Similares em SciELO
Compartilhar
Ingeniería Mecánica
versão On-line ISSN 1815-5944
Resumo
DE SOUZA-ANTUNES, Ricardo; DOS SANTOS-LOUREIRO, Felipe; ANTONIO-SCOLA, Luís e VIEIRA-DE-VASCONCELLOS, João Flávio. Optimization of pipe networks by genetic algorithm employing the colebrook correlation. Ingeniería Mecánica [online]. 2019, vol.22, n.1, pp. 49-56. ISSN 1815-5944.
This paper presents an optimization procedure for identifying the minimum cost of water pipe networks considering a table of commercial diameters. To this end, a real-coding Genetic Algorithm (GA) with the use of a simulated binary and convex crossover and mutation per variable; as well as a dynamic penalty strategywasdeveloped. A computer program to solve the hydraulic model based on the Newton-Raphson method wasdevelopedfor calculating the head loss using Hazen-Williams (HW) and Colebrook correlations. By analyzing a benchmark pipe network example, it is shown that different results are obtained by the HW and Colebrook correlations. Moreover, when simulating the best HW pipe network configuration with the Colebrook correlation, some constraints of the design are violated, indicating that the Colebrook formulation is more adequate to be used in conjunction with the GA due to the randomness of the GA with respect to the Reynolds numbers.
Palavras-chave : pipe networks; hazen-Williams; colebrook; optimization; genetic algorithm.