基于油藏不确定性的适应性井网生成及优化方法研究  

Construction and optimization method of adaptive well pattern based on reservoir uncertainty

作　　者：张凯[1] 张浩 张黎明[1] 姚军[1] ZHANG Kai;ZHANG Hao;ZHANG Liming;YAO Jun(College of Petroleum Engineering, China Petroleum University, Qingdao, Shandong 266580, China)

机构地区：[1]中国石油大学(华东)石油工程学院,山东青岛266580

出　　处：《中国科技论文》2017年第21期2438-2444,共7页China Sciencepaper

基　　金：国家自然科学基金资助项目(61573018;51504277);山东省自然科学基金资助项目(ZR2015EL014);国家高技术研究发展计划(863计划)资助项目(2013AA09A215)

摘　　要：为了提高井网对复杂不确定性油藏的开发效率,针对传统井网存在结构简单,变化不灵活,适应性差等缺点,提出了1种新型的自适应性井网构建方法,利用Delaunay三角形剖分与Frontal-Delaunay四边形剖分法生成大规模高质量的三角形及四边形井网,能够有效地约束各种复杂边界,并实现变密度调整,体现出良好适应性。基于油藏建模存在不确定性,以油田的净现值(net present value,NPV)的期望值和标准差为目标函数,利用PSO算法与k聚类算法相结合,对不确定性油藏模型进行聚类,并循环优化,在保证计算精度的同时,提高收敛效率。经不确定性油藏实例验证,利用所提方法生成的井网能够充分适应油藏的非均质性及复杂边界条件,在降低计算量的同时保证结果精确性,显著降低不确定性带来的开发风险,提高预期收益。To improve the development efficiency of well pattern for complex reservoirs with uncertainties,in view of the shortcomings of traditional well pattern,such as simple structure,inflexible adjustment and poor adaptability,this paper presents a new adaptive well pattern construction method.By using Delaunay Triangulation and Frontal Delaunay Quadrangular division methods,large scaled high quality triangular and quadrangular well patterns are generated,which can effectively deal with various complex boundaries and variable density adjustment,reflecting perfect adaptability.Taking the uncertainty of reservoir modeling into consideration,the expectation and standard deviation of the net present value(NPV)are used to construct the objective function;the PSO algorithm combined with the k clustering algorithm are used to cluster the uncertain reservoir models into groups and then optimize the objective function circularly.In ensuring the accuracy of the calculation,the convergence time is greatly improved at the same time.Examined by the example of uncertainty reservoir,the well pattern generated by this method can fully adapt to the heterogeneity and complex boundary conditions of the reservoir.The proposed optimization method can remarkably reduce the computational complexity while ensuring the accuracy of the results.The development risk caused by the uncertainty is reduced significantly,while the expected revenue is improved correspondingly.

关 键 词：适应性井网 非结构化网格 PSO算法 k聚类算法 不确定性优化 

分 类 号：TE324[石油与天然气工程—油气田开发工程]