• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文学位 >New methods for accurate upscaling with full-tensor effects.
【24h】

New methods for accurate upscaling with full-tensor effects.

机译：具有全张量效果的精确升频的新方法。

获取原文
获取原文并翻译 | 示例
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Upscaling is commonly applied to generate practical reservoir simulation models from highly detailed geocellular descriptions. It is often the case that fine-scale geological features, or the principal correlation directions of the geostatistical model, are not aligned with the simulation grid. For such systems, full-tensor effects generally arise at the coarse scale, even if the fine-scale permeability is isotropic.;In this thesis, new upscaling procedures designed to accurately capture full-tensor effects are developed and evaluated. These techniques are based on variable compact multipoint (VCMP) flux approximations, and are applied to both Cartesian and non-Cartesian grids. The new upscaling procedures generate coarse-scale transmissibilities directly. The inclusion of global flow effects in upscaling computations is known to improve coarse-grid accuracy for highly heterogeneous systems. For this reason, approaches for incorporating global flow effects into the upscaled models are investigated. These include global methods, in which global fine-scale flow information is used for the upscaling, and local-global techniques, in which the global flow information derives from coarse-scale simulations.;We first consider global upscaling methods and develop two procedures within the context of VCMP -- one in which the upscaled model is determined directly (VCMP-DG) and one in which iteration of the coarse-scale model is used to minimize the mismatch between coarse-scale fluxes and integrated fine-scale fluxes (VCMP-IG). These two approaches use a complementary local flow in addition to a fine-scale global flow in the determination of upscaled transmissibilities. VCMP generates locally varying stencils that are optimized for flow accuracy and minimum stencil width. To guarantee monotonicity, the VCMP stencils are adapted to assure the coefficient matrix is an M-matrix whenever nonmonotone solutions are encountered. This is referred to as a selective M-fix procedure.;The new global VCMP methods are applied to multiple realizations of two-dimensional fine-scale permeability descriptions for coarse models defined on both Cartesian and irregular quadrilateral grids. Both log-normally distributed permeability fields with oriented layers and channelized models are considered. Six different upscaling techniques (extended local, direct global, and iterated global, each using both two-point and VCMP flux approximations) are assessed for four different sets of global boundary conditions. The global VCMP techniques consistently display high degrees of accuracy for both pressure and flux. For the oriented-layer cases, where full-tensor effects are important, the global VCMP methods are shown to provide clearly better overall accuracy than analogous methods based on two-point flux approximations. For channelized cases in which full-tensor effects are not significant, both types of methods provide high levels of accuracy. The selective M-fix procedure is also shown to lead to improved accuracy, which can be significant in some cases. In total, for the systems considered here, the new global VCMP upscaling techniques are observed to provide the best overall accuracy of any of the upscaling methods investigated.;Global upscaling methods are not always appropriate because they require global fine-scale flow solutions. Therefore, we also develop and evaluate a variable compact multipoint adaptive local-global technique (VCMP-ALG), as a more efficient alternative to global VCMP methods. This approach avoids global fine-scale computations. The VCMP-ALG method successfully combines the positive attributes of its two underlying component procedures -- the VCMP flux scheme and adaptive local-global (ALG) upscaling. The performance of the local-global VCMP upscaling technique is evaluated for multiple realizations of oriented variogram-based models and synthetic deltaic systems. Extensive numerical results for 2D cases demonstrate that the VCMP-ALG approach provides better overall accuracy than either of the underlying methods applied individually. However, as would be expected, it does not achieve the level of accuracy of the global VCMP methods. We also present results for two-phase oil-water flows and demonstrate that the VCMP-ALG transmissibilities, although computed from single-phase flow computations, are well-suited for use in two-phase flow simulations.;The global VCMP and VCMP-ALG methods described above are also applied to irregular quadrilateral grids. A level of accuracy comparable to that achieved for Cartesian grids is observed. These computations, however, are all for logically Cartesian grids (i.e., grids that maintain a logical i, j structure). In the final portion of this thesis, the upscaling procedures are extended to treat corner-point grids with pinch-outs (in which case the i, j structure is lost). Such grids are often used in practice for modeling geological layers that merge into other layers. Coarse-scale simulation results demonstrate that high degrees of accuracy are again achieved through use of VCMP-ALG or global VCMP methods.
机译：放大通常用于从高度详细的地细胞描述中生成实用的油藏模拟模型。通常情况下，精细尺度的地质特征或地统计学模型的主要相关方向未与模拟网格对齐。对于这样的系统，即使细尺度的渗透率是各向同性的，全张量效应通常也会出现在粗糙尺度上。在本论文中，开发并评估了旨在精确捕获全张量效应的新的放大程序。这些技术基于可变紧凑多点（VCMP）通量近似值，并应用于笛卡尔网格和非笛卡尔网格。新的放大程序直接产生了粗糙的透射率。众所周知，在升尺度计算中包括全局流效应可以提高高度异构系统的粗网格精度。因此，研究了将整体流动效应纳入高级模型的方法。这些方法包括使用全局精细流信息进行升级的全局方法和使用全局流信息从粗略模拟获得的局部全局技术。;我们首先考虑全局扩展方法并在其中开发两个程序VCMP的上下文-一种是直接确定放大模型（VCMP-DG），另一种是使用粗规模模型的迭代来最大程度地减小粗尺度通量和积分精细尺度通量之间的不匹配（VCMP） -IG）。在确定放大的透射率时，这两种方法除了使用精细的全局流外，还使用互补的局部流。 VCMP生成局部变化的模板，这些模板针对流量精度和最小模板宽度进行了优化。为了保证单调性，每当遇到非单调解时，将VCMP模板调整为确保系数矩阵为M矩阵。这称为选择性M固定过程。新的全局VCMP方法应用于在笛卡尔和不规则四边形网格上定义的粗糙模型的二维精细尺度渗透性描述的多个实现。考虑了具有定向层的对数正态分布渗透率场和通道化模型。针对四组不同的全局边界条件，评估了六种不同的放大技术（扩展的局部，直接全局和迭代全局，每种都使用两点和VCMP通量近似）。全局VCMP技术始终显示压力和流量的高精度。对于全张量效应很重要的定向层情况，与基于两点通量近似的类似方法相比，全局VCMP方法显示出明显更高的整体精度。对于全张量影响不明显的通道化情况，这两种类型的方法都提供了很高的准确性。选择性的M-fix程序也显示出提高的准确性，这在某些情况下可能很重要。总体而言，对于此处考虑的系统，观察到新的全局VCMP放大技术可提供所研究的任何放大方法中最佳的总体精度。全局放大方法并不总是合适的，因为它们需要全局精细流解决方案。因此，我们还开发和评估可变紧凑多点自适应局部全局技术（VCMP-ALG），作为全局VCMP方法的更有效替代方法。这种方法避免了全局精细计算。 VCMP-ALG方法成功地结合了其两个基本组件过程的积极属性-VCMP通量方案和自适应局部全局（ALG）放大。针对定向变函数图为基础的模型和合成三角洲系统的多种实现，评估了局部全局VCMP放大技术的性能。二维案例的大量数值结果表明，与单独应用的两种基础方法相比，VCMP-ALG方法可提供更好的总体准确性。但是，不出所料，它无法达到全局VCMP方法的准确性。我们还提供了两相油水流的结果，并证明了VCMP-ALG的透射率，尽管是从单相流计算中得出的，却非常适合用于两相流模拟中。全局VCMP和VCMP-上述ALG方法也适用于不规则四边形网格。可以观察到与笛卡尔网格可比的精度水平。但是，这些计算全部用于逻辑上的笛卡尔网格（即，保持逻辑i，j结构的网格）。在本文的最后部分，扩展程序被扩展为处理带有收缩的角点网格（在这种情况下，，j结构丢失）。实际上，这种网格通常用于对合并到其他层中的地质层进行建模。粗规模的仿真结果表明，通过使用VCMP-ALG或全局VCMP方法，可以再次实现较高的精度。

著录项

  • 作者

    Chen, Tianhong.;

  • 作者单位

    Stanford University.;

  • 授予单位 Stanford University.;
  • 学科 Engineering Petroleum.
  • 学位 Ph.D.
  • 年度 2009
  • 页码 118 p.
  • 总页数 118
  • 原文格式 PDF
  • 正文语种 eng
  • 中图分类 石油、天然气工业;
  • 关键词

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号