首页> 外文学位 >Structured flow-based gridding and upscaling for reservoir simulation.
【24h】

Structured flow-based gridding and upscaling for reservoir simulation.

机译:基于流量的结构化网格划分和放大,用于油藏模拟。

获取原文
获取原文并翻译 | 示例

摘要

When we construct coarse models for reservoir simulation based on fine grid geocellular descriptions, the use of curvilinear grids enables more efficient representations of important features in the reservoir. Upscaling methods compatible with these grids must also be applied.; In this thesis, we develop and apply gridding and upscaling techniques in the context of structured grids. In 2-D, for grid generation techniques, we first considered streamline-based gridding with Laplacian smoothing to capture high flow regions effectively and with improved grid quality over pure streamline grids. Two other structured gridding methods (the Jacobian-based elliptic method and optimization-based method) were investigated to achieve better local control of grids and incorporate additional information (not necessarily flow-based). Upscaling techniques to compute effective properties for these grids were also implemented. We first developed a new finite element-based method to resolve the coarse cell geometry accurately. We also developed two finite difference-based procedures (permeability upscaling and transmissibility upscaling). They offer an approximate treatment of the coarse cell geometry but display better computational efficiency than the finite element method. We observed similar accuracy in 2-D flow results for the finite element and the finite difference-based methods. We therefore implemented the more efficient finite difference methods in 3-D. The gridding techniques implemented in 3-D include the streamline-based method with Laplacian smoothing and the Jacobian-based elliptic method. We extended permeability upscaling and transmissibility upscaling to 3-D and compared their performance (within the context of multipoint and two-point flux approximations, respectively) for several cases.; We evaluated our gridding and upscaling techniques for various heterogeneous permeability fields and flow scenarios. These flow tests demonstrate that for many cases, the coarse models constructed by combining our gridding and upscaling methods result in considerable improvements over uniform Cartesian models at the same coarsening level. This demonstrates that our gridding procedures correctly resolve the critical features in the reservoir and that our upscaling techniques introduce appropriate effectivizations of subgrid heterogeneity. The tests also show that transmissibility upscaling generally produces better results than permeability upscaling.
机译:当我们基于精细的网格地细胞描述构建用于储层模拟的粗略模型时,使用曲线网格可以更有效地表示储层中的重要特征。还必须采用与这些网格兼容的升级方法。在本文中,我们在结构化网格环境中开发并应用了网格化和放大技术。在二维中,对于网格生成技术,我们首先考虑了基于流线的网格化和Laplacian平滑,以有效捕获高流量区域并具有比纯流线网格更高的网格质量。研究了另外两种结构化网格化方法(基于Jacobian的椭圆方法和基于优化的方法),以实现对网格的更好的局部控制并合并其他信息(不一定基于流)。还实现了为这些网格计算有效属性的升级技术。我们首先开发了一种新的基于有限元的方法来精确解析粗胞几何。我们还开发了两种基于有限差分的程序(渗透率提升和透射率提升)。它们提供了粗单元几何的近似处理,但显示出比有限元方法更好的计算效率。对于有限元和基于有限差分的方法,我们在二维流动结果中观察到了相似的精度。因此,我们在3-D中实现了更有效的有限差分方法。在3-D中实现的网格化技术包括基于流线的Laplacian平滑方法和基于Jacobian的椭圆方法。我们将渗透率提升和透射率提升扩展到了3-D,并在几种情况下比较了它们的性能(分别在多点和两点通量近似的背景下)。我们针对各种非均质渗透率场和流动情景评估了网格化和放大技术。这些流量测试表明,在许多情况下,结合我们的网格划分方法和放大方法构建的粗略模型比在相同粗略水平下的统一笛卡尔模型产生了显着改进。这表明我们的网格化程序正确地解决了油藏中的关键特征,并且我们的升级技术引入了亚网格异质性的适当实现。这些测试还表明,传输率提升通常比渗透率提升产生更好的结果。

著录项

  • 作者

    He, Chuanping.;

  • 作者单位

    Stanford University.;

  • 授予单位 Stanford University.;
  • 学科 Engineering Petroleum.
  • 学位 Ph.D.
  • 年度 2005
  • 页码 191 p.
  • 总页数 191
  • 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