基于遗传算法的气田集输管网整体优化方法  

A global optimization method based on genetic algorithms for gas gathering pipeline network in a gas field

作　　者：李自力[1] 孙云峰[1] 张子波[2] 刘静 

机构地区：[1]中国石油大学(华东)储运与建筑工程学院 [2]北京油气调控中心 [3]北京中海石油研究中心

出　　处：《天然气工业》2011年第8期86-89,137,共4页Natural Gas Industry

基　　金：国家科技重大专项"大型油气田及煤层气开发--高含硫气田集输工艺与安全控制技术"项目资助(编号:2008ZX05017-004)

摘　　要：目前气田常用天然气集输管网分级优化法的缺点在于每一级的优化结果仅仅是局部最优解,难以保证最终结果是全局最优解。为此,提出了基于遗传算法的气田集输管网整体优化方法,即以整个管网的投资费用最低为目标函数建立数学模型,采用遗传算法求解,在遗传算法中将各站的站址及站辖管道的管径放在一起作为优化变量,求出管网投资费用最低时各站的站址以及各条管道的管径;同时针对遗传算法本身的局限性,结合模拟退火算法,调整了优化方法的适应度函数。实例计算结果表明,该整体优化方法在节省投资方面要优于传统的分级优化方法。The commonly used hierarchical optimization method for the field gas gathering pipeline network can only ensure the hierarchical optimization results, but not the global optimization result. This paper hereby presents a global optimization method based on genetic algorithms for the gas gathering pipeline network in a gas field. Firstly, a mathematical model is established, in which the minimal investment of a whole network is taken as the objective function. Secondly, in genetic algorithms, the site of each station together with the diameter of each pipe within each station is taken as optimization variables, thus the station site and the diameter of each pipe can be solved. Finally, aiming at the limits of the genetic algorithms, the simulated annealing method is also used to adjust the fitness function of the optimization method. Case studies show that this proposed global optimization method based on genetic algorithms is superior to the traditional hierarchical optimization method especially in saving the investment cost.

关 键 词：天然气集输管网 遗传算法 整体优化 模拟退火 投资费用 优化变量 

分 类 号：TE863[石油与天然气工程—油气储运工程]