Genetic Algorithms Applied to the Optimum Design of Gas Transmission Networks S.J. Montoya-O.; S.J. Montoya-O. Universidad Nacional de Colombia Search for other works by this author on: This Site Google Scholar W.A. Jovel-T.; W.A. Jovel-T. Universidad Nacional de Colombia Search for other works by this author on: This Site Google Scholar J.A. Hernandez-R.; J.A. Hernandez-R. Universidad Nacional de Colombia Search for other works by this author on: This Site Google Scholar C. Gonzalez-R. C. Gonzalez-R. Universidad Nacional de Colombia Search for other works by this author on: This Site Google Scholar Paper presented at the SPE International Petroleum Conference and Exhibition in Mexico, Villahermosa, Mexico, February 2000. Paper Number: SPE-59030-MS https://doi.org/10.2118/59030-MS Published: February 01 2000 Cite View This Citation Add to Citation Manager Share Icon Share Twitter LinkedIn Get Permissions Search Site Citation Montoya-O., S.J., Jovel-T., W.A., Hernandez-R., J.A., and C. Gonzalez-R. "Genetic Algorithms Applied to the Optimum Design of Gas Transmission Networks." Paper presented at the SPE International Petroleum Conference and Exhibition in Mexico, Villahermosa, Mexico, February 2000. doi: https://doi.org/10.2118/59030-MS Download citation file: Ris (Zotero) Reference Manager EasyBib Bookends Mendeley Papers EndNote RefWorks BibTex Search Dropdown Menu nav search search input Search input auto suggest search filter All ContentAll ProceedingsSociety of Petroleum Engineers (SPE)SPE International Oil Conference and Exhibition in Mexico Search Advanced Search AbstractThis paper describes the application of a modified genetic algorithm (GA) to optimize the design of gas transmission networks operating under steady-stable conditions.The gas-network optimization system is implemented by means of a genetic algorithm that executes an "intelligent search" in a "gas transmission networks" space.It obtains automatically and with a low computational work a distribution of pipeline segment diameters for a minimum investment cost, accomplishing some restrictions initially fixed.IntroductionThe design of a new gas pipeline transmission system or the expansion of an existing one requires the optimization of its topology, size and operational conditions to minimize investment and maintenance cost. Traditional solutions to this problem use gradient-based techniques. An acceptable convergence for these methods depends on the initial values given by the designer.Nowadays, there exists a new problem-solution paradigm called Evolutionary Computation, which could also find applicability to solve petroleum-engineering problems. The Genetics Algorithms (GA) are part of these techniques. GA are searching methods that imitate the mechanisms observed in the Darwinian Theory of Natural Selection of Species. GA are massive parallel algorithms which start with a known or unknown set of possible solutions for the target problem. Genetic operators are randomly applied to this set to obtain a new potential population of solutions. This process is repeated until reaching an optimal solution or a global optimal near solution. The only input for GA are the features of the target problem and the needed cost function.This paper presents a GA application to the optimal design of a steady-state gas transmission. The studied case consists of a network operating under normal conditions in the low-middle pressure range and twenty-four commercial pipeline segments. The GA developed in this study has already been applied satisfactorily to optimize a network built in the Mexico valley 1, obtaining a significant decrease in the investment cost.Gas Transmission Network DesignThe gas transmission networks are systems compound for a series of valves, pipes, bombs, etc., connected in such a way that they give the consumers the quantities of demanded gas at the required pressures.The gas transmission network design is a complex problem because discret parameters are used. This is a non-lineal and multimodal problem what increases its complexity yet for networks of moderate size.Conventionally, the networks have been designed using iterative methods like Hardy-Cross, Newton-Raphson and linear theory, among others.The proposed model in this study integrates the traditional methods of gas networks solution with a modified GA. This model allows one to optimize and to design the gas networks under stable flow conditions.Genetic AlgorithmsGen and Cheng 2 define the GA as a powerful and broadly applicable stochastic search and optimization techniques based on principles from evolution theory. Keywords: colombia, gas transmission network, optimization, fitness, chromosome, spe 59030, selection, diameter, constraint, evolutionary algorithm Subjects: Information Management and Systems, Artificial intelligence This content is only available via PDF. 2000. Society of Petroleum Engineers You can access this article if you purchase or spend a download.