摘要
我国海上油田原油资源储量丰富,其中N油田有着大量的稠油资源,现有的多元热流体开采方式已无法进一步提高油田采收率,开采效果差,因此急需开展蒸汽驱注采参数优化研究,但是传统的优化方法主要针对不同参数逐一进行优化方法,无法从全局综合考虑各个参数的协同优化作用,因此最终方案往往是单一参数的最优方案而不是全局最优方案。首先利用Petrol建立了区块地质模型,进而采用数值模拟软件CMG对该油田试验区进行了生产历史拟合,在此基础上使用基于均匀设计的改进型粒子群优化算法(PSO算法)对蒸汽驱开采方案进行整体智能优化,并将预测的开发效果与传统优化方案进行对比。对比结果显示,使用智能优化方案对油田进行开采,累增油量34.05万方,提高采收程度12.8%,相对人工配产方案增油5.39万方,提高采收率2.03%,有着更佳的开采效果。
The offshore oilfields are rich in crude oil resources,of which N Oilfield has a large number of heavy oil resources.The existing multiple thermal fluid extraction methods have been unable to further improve oilfield recovery factor,and the recovery effect is poor,so it is urgent to carry out steam flooding injection and production parameter optimization research,but the traditional optimization method mainly focuses on the optimization of different parameters one by one.Unable to considering the coordinated optimization of each parameter,the final solution is often the optimal solution with a single parameter rather than the global optimal solution.This article first established a block geological model using Petrol,and then used the numerical simulation software CMG to fit the production history of the oilfield test area.Based on this,an improved particle swarm optimization algorithm(PSO algorithm)based on uniform design was used for steam.The overall development of the mining and mining scheme is optimized intelligently,and the predicted development effect is compared with the traditional optimization scheme.The comparison results show that the use of intelligent optimization schemes for oilfield extraction has cumulatively increased the amount of oil by 340,500 cubic meters,increased the recovery by 12.8%,compared with the artificial production plan,the increase of oil amount was 53,900 cubic meters,and the recovery factor was increased by 2.03%,has better extraction results.
作者
孙玉豹
王少华
吴春洲
肖洒
SUN Yu-bao;WANG Shao-hua;WU Chun-zhou;XIAO Sa(China Oilfield Services Limited. (COSL Production Optimization) Tanggu Marine High-tech Development Zone,Tianjin 300459,China)
出处
《科技和产业》
2020年第11期221-228,256,共9页
Science Technology and Industry
基金
国家重点研发计划(2018YFA0702400)。
关键词
蒸汽驱
数值模拟
注采参数优化
无梯度智能优化
历史拟合
steam flooding
numerical simulation
optimization of injection and production parameters
intelligent optimization without gradients
history fitting
