Fractured Reservoir Simulation via Multi-Scale Finite Volume
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for simulating fluid flow in fractured porous media, such as dual porosity and discrete fracture modeling, are inefficient and unsuitable for realistic scenarios due to high computational costs and complexity, especially in dynamic fracture networks and highly conductive fractures.
Innovation Solution
A method that uses a hierarchical fracture model with multi-scale finite volume (MSFV) techniques to construct and sequentially solve systems of equations based on scale separation, incorporating fracture and matrix equations coupled via flux interactions, allowing for accurate and efficient simulation of fluid flow in fractured subterranean reservoirs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If discrete fracture modeling with complex unstructured gridding is used to accurately capture fracture geometries, then measurement precision of fracture characteristics is improved, but device complexity and computational cost increase significantly
Solution Approach 1:
The domain is segmented into matrix blocks and fracture networks separately. The matrix is discretized into blocks that may contain multiple fractures, while fractures are represented as separate entities with their own geometry. This segmentation allows accurate fracture geometry capture without requiring complex global grid conforming to every fracture detail.
Solution Approach 2:
Fractures are extracted from the matrix domain and represented as separate discrete entities. Instead of embedding fracture geometries into the matrix grid, fractures are taken out as independent features with their own properties, allowing simplified matrix gridding while maintaining accurate fracture representation.
2Measurement precision
If fine-scale simulations are used to capture high contrast physical properties and length scales, then measurement precision of flow characteristics is improved, but use of energy and computational cost increase
Solution Approach 1:
Different parts of the domain are treated with different levels of detail. The matrix uses a coarser grid for computational efficiency, while fractures use their own discrete representation for accurate geometry capture. This local differentiation allows fine-scale flow characteristics to be captured in fractures without requiring fine-scale resolution throughout the entire domain.
Solution Approach 2:
The problem is solved by transitioning from a fully three-dimensional fine-scale simulation to a reduced-dimensional representation where fractures are treated as separate entities with their own geometry and flow equations. This dimensional separation maintains accuracy while reducing computational cost.
3Device complexity
If dual porosity model with upscaling strategy is used to simplify fracture representation, then device complexity is reduced, but adaptability to long scale fractures deteriorates
Solution Approach 1:
The model dynamically adapts its representation based on fracture characteristics. Fractures are represented as discrete entities with explicit geometry, allowing the model to naturally handle various fracture scales from small to long scale without requiring different model formulations. This dynamic representation maintains adaptability across different fracture configurations.
4Measurement precision
If complex conforming grids are generated to capture fracture geometries accurately, then measurement precision of fracture characteristics is improved, but ease of manufacture and implementation deteriorates
Solution Approach 1:
The grid generation process is segmented into independent matrix block creation and fracture network representation. Matrix blocks are generated using simple, standard grid generation techniques, while fractures are added as separate discrete entities. This segmentation eliminates the need for complex conforming grid generation to capture fracture geometries.
Solution Approach 2:
Fracture geometry capture is extracted from the matrix grid generation process. Instead of requiring the matrix grid to conform to fracture geometries, fractures are represented as separate entities with their own geometry, decoupling the grid generation complexity from the fracture representation accuracy.
Data Source
AI summary
A method, system and computer program product are disclosed for simulating fluid flow in a fractured subterranean reservoir. A reservoir model representative of a fractured subterranean reservoir is provided. The reservoir model includes porous matrix control volumes and a network of fractures, which define fracture control volumes, overlying the porous matrix control volumes. A system of equations based on scale separation is constructed for fluid flow in the porous matrix control volumes and the fracture control volumes. The system of equations can include fracture equations having a pressure vector for each network of fractures that is split into an average pressure value and remainder pressure value. The system of equations based on scale separation is sequentially solved, such as by using an iterative Multi-Scale Finite Volume (MSFV) method.