Browsing by Subject "CFD-DEM"
Now showing 1 - 2 of 2
- Results Per Page
- Sort Options
Item Injection-induced fractures and polymer injectivity in unconsolidated formations : fundamentals, experiments and modeling(2022-11-17) Li, Zihao, Ph. D.; Espinoza, David N.; Balhoff, Matthew T.; Mohanty, Kishore K.; Delshad, Mojdeh; Mondal, SomnathPolymers are often used for enhanced oil recovery (EOR) and in-situ site remediation. Polymer injectivity is an essential indicator in the project economics. Reduced injectivity due to the high polymer viscosity can be one of the most important challenges in a field polymer flood project, more so than the cost of chemicals. Although injection of viscous polymer is expected to significantly reduce the injectivity, observations from field data of polymer floods often show higher than expected injectivity for viscoelastic HPAM polymers. One explanation for the unexpected injectivity is that the injection pressure may increase above the formation parting pressure (FPP) at which fractures are created from the wellbore. The effects of non-Newtonian rheology on fracture opening and fluid leakoff morphology is largely unexplored compared to Newtonian fluids. Comprehensive studies that explain the polymer-driven fracturing mechanisms are lacking, even in unconsolidated sands. In this study, we study polymer-driven fracture initiation through experiments and numerical simulations. First, we conduct experiments injecting Newtonian, shear-thinning and viscoelastic fluids into a sand-filled Hele-Shaw cell to study fluid invasion and fracturing patterns of sand and injected fluids. The experimental results demonstrate complex fracture patterns depending on injection velocity, fluid viscosity, boundary conditions and fluid rheology. We use digital image correlation to demonstrate that the fracturing mechanism in sand is associated with shear dilation at the tip of fracture and fracture offshoots. Compared to Newtonian fluids, shear-thinning fluids cause thinner fracture openings and more leakoff. Injecting viscoelastic fluid into sand creates a more irregular leakoff pattern, with fluid sometimes leaking mostly at the fracture tip for high injection rates. Second, we investigate the mechanisms behind polymer-driven fractures in cohesionless granular media by coupling the discrete element method (DEM) with computational fluid dynamics (CFD) at the pore/grain scale. Fluid injection can initiate a fracture orthogonal to the minimum principal stress, as expected, through viscous drag forces. However, the displacement field around the fracture is much more complex than what is expected for a linear elastic solid, showing that (1) fracture propagation requires small leak-off (a few times the fracture width) and (2) localized shear deformation ahead of the fracture tip affects fracture morphology. Finally, we simulate the injection of the polymer into the unconsolidated sand. Polymer injection can create fractures in the granular media along the direction perpendicular to the minimum principal stress, thereby reducing wellbore pressure buildup at a constant polymer injection rate. Polymer rheology, water quality, and undissolved polymer also affect the polymer injectivity. This work shows the initiation of polymer-driven fractures in a granular model and demonstrate its implications on polymer injectivity.Item Modeling particulate flows in conduits and porous media(2018-12) Wu, Chu-Hsiang; Sharma, Mukul M.; Lake, Larry; Mohanty, Kishore; Prodanovic, Masa; Mondal, SomnathParticulate flows exist in a wide range of engineering applications such as drilling and completion operations—from hole-cleaning to well stimulation to sand control. The operations involving particulate flows typically focus on achieving either (1) efficient particle transport or (2) effective particle retention, both of which require a deep understanding of particulate flow behavior under different flow conditions. This dissertation presents novel approaches for modeling particulate transport with the intent to optimize operational efficiency in specific important oilfield applications. The first half of this dissertation focuses on modeling particle transport inside a wellbore. A typical example of such an operation is the pumping of proppant/diverter slurry during a hydraulic fracturing treatment. During the treatment, the particles are moved in a carrier fluid from the wellbore through the perforations and finally into the fractures. The motion of the particles is primarily influenced by the interactions between the fluid phase and the solid phase, and therefore, a precise description of the particle-fluid interactions is essential for modeling the process. The coupling of computational fluid dynamics with the discrete element method (CFD-DEM) approach is adopted for this task. The investigation begins with evaluating the particle transport efficiency (PTE) through a perforation in a horizontal wellbore under various downhole flow conditions. A comprehensive study on the effect of casing, perforation, solid, and fluid properties on PTE is presented. On the basis of the study, empirical PTE correlations are derived and integrated into a multi-cluster hydraulic fracturing model to simulate proppant/diverter transport at a wellbore scale. Simulation results show that because of its high inertia, the transport of proppant is generally much more difficult than the transport of carrier fluid through the perforations. By assuming a simple jamming criterion based on local proppant concentration near the perforations, the heel-biased treatment distribution commonly observed in the field can be accurately reproduced by the model. Recommendations on fracturing job design for promoting an even treatment distribution are also discussed. The second half of the dissertation focuses on modeling particle retention in sand control completions with a special focus on multi-layered metal-mesh screens and gravel packs. A DEM-based approach is developed to evaluate the pore throat size distribution (PoSD) of the sand-retention media. Simulation results show that for a multi-layer plain square mesh (PSM) screen, the overlap, alignment, and relative pore size between individual layers all have a significant impact on the screen’s PoSD. In contrast, only the intra-layer overlap of the filter is important for controlling the PoSD of a multi-layered plain Dutch weave (PDW) screen. For gravel packs, simulation results show that the largest and smallest pore throat sizes are about 1/5 and 1/10 of the effective gravel size for typical gravel sizes used in the field. By using the computed PoSD as an input, an analytical model and a Monte-Carlo model are developed to predict sand production through gravel packs. The modeled PoSD and sand production agree reasonably well with field observations and experimental data. Our approach enables completion engineers to tailor gravel pack designs for different combinations of formation sand size distribution (PSD) and wellbore geometry in a cost-efficient manner.