点坝砂体内部非均质性的层次建模法

Fluvial reservoirs are an important kind of deposition systems as well as an important kind of reservoirs. However, characterization and modeling of fluvial reservoirs are challenging because of the various scales of heterogeneity that exist between and within fluvial deposits. Point-bar, a very important part in meandering channel reservoir, is more difficult to characterize with subsurface data because of the relatively limited lateral extent of point-bar deposits, which are commonly less than the typical well spacing in developed fields. Point-bar deposits are formed by lateral channel migration and can be divided into abandoned channel, lateral-accretion units and associated shale drapes. The size, proportion and extending length of these shale drapes can influence reservoir behavior, because they are potential baffles and barriers to fluid flow. In addition, internal facies variations, such as the vertical change from trough cross-bedded to ripple laminated sandstone, produce a fining-upward trend in grain size and a corresponding decrease in porosity and permeability. These features increase the difficult of characterization and modeling of point-bar. .We propose a new geostatistical modeling framework of point-bar deposition system. Firstly, the characterization and forming conditions of heterogeneity within point-bar are invested based on the research of literatures, outcrops, satellite images and so on. According to the heterogeneity of point-bar is hierarchical, the architecture elements within point-bar are modeled firstly. Under the control of architecture elements model, the porosity and pemerability models are builded. A new modeling algorithm will be developed based on Alluvism-a process-based modeling algorithm. It will be used to construct a 3D training image to characterize the distribution of architecture elements within point-bar. Multiple-point geostatistics will be used to build a 3D architecture model of Gudao oilfield based on the spatial correlations referred from the previous builted 3D training image. Facies-control modeling method under trends constraint will be used to simulate petrophysical properties. The models will be checked by geological integration analysis and stream line simulation to improve the modeling algorithm. The new method can explicitly model lateral-accretion units and associated shale drapes, in addition, more flexible petrophysical property trends in lateral-accretion are considered. The prediction of remaining oil will benefit from the geological models constructed by the new method. This project will be of an improtant meaning for enhancing oil recovery.

点坝砂体内部非均质性严重，对剩余油分布具有重要的控制作用，现有的建模方法在定量表征点坝内部非均质性方面存在不足。本研究拟通过文献调研、露头解剖、卫星图片解释与测量、水槽沉积模拟等多种手段深化点坝内部非均质性形成机理及定量描述方面的研究。针对点坝内部非均质性具有层次性的特点，首先模拟点坝内部构型单元的分布，刻画泥岩夹层等构型单元导致的非均质性；在构型单元模型基础上模拟物性参数在砂体内部的分布，进而刻画次一级的非均质性特征。在模拟方法上，拟发展新的基于沉积过程的建模算法以建立表征点坝构型分布的三维原型模型；以原型模型为基础，采用多点地质统计学方法建立点坝构型的三维模型；利用趋势约束的相控建模技术建立储层物性参数模型。拟对孤岛油田进行实例检验，完善建模算法。项目的实施，将进一步深化点坝内部非均质性形成机理方面的认识，形成点坝砂体内部非均质性三维定量表征的建模方法，对于剩余油的开发具有重要的意义。

曲流河储层是重要的一类油气储层，由于其砂体内部非均质性复杂，导致剩余油预测难度大，而现有的建模方法难以合理刻画砂体内部结构，因此开展了本项目的研究，形成新的建模技术。.项目主要开展了以下几方面的研究：1）对基于沉积过程的建模算法进行了深入剖析，针对已有建模算法无法刻画点坝砂体内部构型的不足，对该算法进行了改进，实现了点坝内部构型的模拟，改进后的算法能够灵活地控制点坝侧积层的倾角、延伸长度、频率等对流体运动起重要作用的关键参数，实现了河道的部分或全部废弃；2）对建模算法的并行化方法进行测试，主要针对多核CPU基于目标算法并行计算的测试，探讨如何利用Visual C和.NET Framework的并行扩展把建模算法从串行化模式改写为并行化模式；3）开展了建模参数敏感性测试，包括克里金插值算法中参与计算的数据多少方面的敏感性分析；影响河流相储层砂体连通体积的不确定性因素的定量评价；4）提出了一种新的模型优选方法。为了从多个模型中快速有效地选出具有代表性的模型，提出了利用K均值聚类分析方法进行模型优选。以每两个模型之间相对应的每一网格节点的属性值差的平方和，构建表征各模型之间差异的一个量化矩阵；应用多维尺度分析技术对矩阵进行降维，采用K均值聚类分析方法对模型进行聚类；5）在基于目标的建模算法的基础上，提出了曲流河点坝内部结构建模的层次法。按照3个不同的层次分别模拟河道、点坝及侧积层。以地质知识库为指导，建立了中一区馆3先导试验区三维点坝侧积层模型，对剩余油的预测提供了坚实的地质基础。通过本项目，形成了一套表征点坝砂体内部非均质性的建模方法和技术，并在先导试验区成功应用，具有良好的应用前景。.研究成果作为重要支撑材料的项目获得湖北省科技进步一等奖。相关研究成果多次在国内外学术会议上交流，产生了较大影响。依托本项目，公开发表了15篇学术论文，获得4项软件著作权，培养的研究生获得省优博士论文1篇、省优硕士论文2篇。