Semianalytical stream-function solutions on unstructured grids for flow in heterogeneous media

Randy D. Hazlett, D. Krishna Babu, Larry W. Lake

Research output: Contribution to journalArticlepeer-review

4 Citations (Scopus)

Abstract

This paper outlines a Boundary Element Method (BEM) for a piece-wise analytic solution of the Laplace (Poisson) equation for pseudosteady-state, single-phase flow on unstructured, rectangular grids. The method models flow through a reservoir that has been segmented into interacting homogeneous rectangular regions; no further discretization of the solution space analogous to grid refinement in numerical schemes is required for improved accuracy. Rather, boundary discretization allows for continuation of pressure and flux. Previous work on pressure distribution modeling is extended to analytically capture the stream function. Stream-function solutions can then form the basis for other performance measures, such as improved oil recovery efficiency estimation or tracer flow analysis. Moving beyond structured grids into unstructured grid geometry allows for advanced flexibility in problem development and improved efficiency in solution construction. The analytic approach avoids the need for numerical differentiation of the pressure field and particle tracking methods to recover streamlines. Capturing flow in highly heterogeneous media, without local grid refinement, is demonstrated to showcase the robustness of the technique in handling complex reservoir architecture, of particular interest in optimal well positioning and optimal well-pattern development.

Original languageEnglish
Pages (from-to)179-187
Number of pages9
JournalSPE Journal
Volume12
Issue number2
DOIs
Publication statusPublished - Jun 2007
Externally publishedYes

ASJC Scopus subject areas

  • Energy Engineering and Power Technology
  • Geotechnical Engineering and Engineering Geology

Fingerprint

Dive into the research topics of 'Semianalytical stream-function solutions on unstructured grids for flow in heterogeneous media'. Together they form a unique fingerprint.

Cite this