Computational fluid dynamics modelling of fluid flow inside fractured reservoirs.

Al-Mashhadanie, Haider Abdulkareem Jasim

Abstract

Fractured media exist in most layers of the earth's crust, often dominating bulk properties of subsurface geological formations. Therefore, fractured media are involved in many key engineering sectors that impact humans living on Earth. Fractured formations consist of two distinct media sharing the same location: matrix and fracture, which affect each other's flow. Both have heterogeneous properties, such as anisotropic matrix permeability and rough fracture surfaces; also, fractures have varied orientation angles and exist in fractured formations in either discrete fracture form, or in connected networks with varied angles/patterns. Due to this heterogeneity, most fractured media modelling and studies in the literature have considered assumptions that don't represent flow in realistic conditions. This thesis therefore presents systematic investigations conducted on fractured media by using Computational Fluid Dynamic ANSYS CFD Fluent FVM to investigate fluid flow in many kinds of fractured media. These investigations ranged from simple and widely used fractured geometries to more complex ones. The research began with parallel plates fractures and rough fractures with horizontal orientation inside fractured domains. Both fractures were investigated with different fracture surface conditions in these fractured domain models to create most mimicked realistic fractured formation conditions; for example, through inclusion and exclusion of the matrix effect on flow, and matrix isotropic/anisotropic permeability's effects on flow. The results of these models were validated and compared with current understandings of fractured media model flow in the literature. The outcomes of these models have reflected that parallel plates fractures with a single aperture are unsuitable for representing flow in fractured media. These investigations have also shown that exclusion of matrix in fractured media flow will highly mislead flow calculations in fractured media. The second part of the research involved using the results of ANSYS CFD Fluent FVM rough fracture models to develop two fracture friction-factor models in realistic fractured media conditions (analytical and numerical friction-factor models). These accounted for the effect on entire fracture domain flow of both rough fracture geometry effects and also matrix permeability. Specifically, isotropic and anisotropic matrix permeability along layers of formations, along fracture length and in two directions of flow, X and Y, considering Kx and Ky anisotropic effect on lateral and perpendicular flow of each layer. Friction-factor is important for predicting pressure drop (Delta P / L) along fractures and accordingly on fluid migration in fractured formations. In the third part of the research, ANSYS CFD Fluent FVM rough fracture network models were created, which included many heterogeneous properties of fracture media, such as many patterns of network orientations (where each model has different inlets and outlets of flow) and matrix effect (including isotropic and anisotropic matrix permeability). The outcomes of these models have resulted in a new and interesting understanding of modelling fractured media. The research has proven that matrix functionality in fractured media is not only as fluid provider, but also has major effects on providing and transporting fluids in fractured media. As well, the research has provided new evidence that modelling a single fracture of fractured media will highly mislead flow calculation.

Citation

AL-MASHHADANIE, H.A.J. 2021. Computational fluid dynamics modelling of fluid flow inside fractured reservoirs. Robert Gordon University, PhD thesis. Hosted on OpenAIR [online]. Available from: https://doi.org/10.48526/rgu-wt-1603660

• Thesis Type: Thesis
• Deposit Date: Feb 25, 2022
• Publicly Available Date: Feb 25, 2022
• Keywords: Fluid dynamics; Fractures; Fracture formations; Environmental engineering
• Public URL: https://rgu-repository.worktribe.com/output/1603660
• Publisher URL: https://doi.org/10.48526/rgu-wt-1603660
• Award Date: Sep 30, 2021