基于遗传算法的集输管网井组划分  

Well Division of Gathering and Transferring Pipeline Networks by Genetic Algorithm

作　　者：吴华丽[1] 陈坤明[1] 王效东[1] 徐武峰[2] 

机构地区：[1]西南石油大学,四川成都610500 [2]中石油陕西销售公司咸阳分公司,陕西咸阳712000

出　　处：《管道技术与设备》2007年第6期1-2,4,共3页Pipeline Technique and Equipment

基　　金：西南石油大学研究生创新基金(cxjj27026)

摘　　要：通过对油田集输管网的分析研究,建立了基于产量距离之和最小的井组最优划分的数学模型,并采用基于整数编码的遗传算法对模型进行了求解。实例计算表明:建立的以产量距离之和最小为目标的数学模型更能反映集油管网投资的经济性,采用基于整数编码的遗传算法进行求解,其寻优效益高,且收敛性和稳定性也较好。最后求得的产量距离和为18937.93 km.t/d,比文献[2]中的划分结果小1381.07 km.t/d.By analyzing and researching the oil gas gathering and transportation pipeline network,taking the shortest total yield distance as the object function,a mathematical molel is established for dividing the group of wells optimally,then it has solved the model by Genetic Algorithm(GA) based on integer coding.The calculation of the examples' shows that the mathematical model with the shortest total yield distance as the object function can better reflect economical value of investment in the oil gathering pipeline;meanwhile,the profit of seeking the optimal result by GA based on integer coding is higher,thus helping obtain a better astringency and stability.As a result,the final shortest total yield distance is 18 937.93 km·t/d,which was 1 381.07 km·t/d shorter than the result of the literature[2].

关 键 词：集输管网 井组划分 优化设计 数学模型 遗传算法 

分 类 号：TE862[石油与天然气工程—油气储运工程]