Abstract
Design and implementation of gas distribution networks is one of the most complex and expensive stages in gas transport projects. Usually there is no optimization on pipes diameter.
Genetic algorithm is based on "Natural Selection" idea of Darwin. Natural selection is the process by which heritable traits that makes it more likely for an organism to survive and successfully reproduce become more common in a population over successive generations. So, any genetic algorithm is composed of an initial population in which bests are chosen to produce next population. This population is subject to mutation and finally next generation is produced which has its parents traits. Theoretically and experimentally evolutionary algorithms are proven to be an efficient and powerful tool in finding optimum solution in a reasonable time. Using genetic algorithms to optimize pipes diameter, the costs of distribution projects can be reduced significantly.
In this study, gas consumption was estimated in a certain area using statistical methods based on maximum gas consumption of commercial units and households. As a result, 5 pipes should have 1.5″ and 1.25 diameters and the rest should have a diameter of 1″. Because of program parameters, the final results will always be in allowable range of velocity and pressure. Present research showed 64% optimization compared to traditional approach. Economically this would be a great saving on expenses.
According to results, final solution is the best solution for this problem in which all standards are observed and genetic algorithm is an efficient tool to find optimum solution of everyday problems. The advantages of this approach are:
Easy Transfer of topology data to program;
Usage of this program to optimize any distribution network;
All formulas, allowable diameters and corresponding cost can be changed readily; and
Fast to find optimum solution, even in large and complex networks.
