A series of hydraulic fracturing injection experiments was performed in unconsolidated sand in a radial flow cell (RFC) to delineate the fundamental mechanisms controlling fracture propagation. The results of this work will be used to optimize the current fracpacking practices in unconsolidated and poorly consolidated sand and to develop adequate modeling techniques. Although the generic word "fracture" is used throughout this document and elsewhere to describe injection-induced formation parting during stimulation, it will be made clear that the tip propagation mechanisms in unconsolidated sand are fundamentally different from linear elastic fracture mechanics (LEFM) as defined for competent rocks. Conceptually, the RFC used for the injection tests simulates a 1-ft thick section of unconsolidated sand formation within a 3-ft radius of a wellbore. The tests in this system included injection of cross-linked guar and visco-elastic surfactant into 3,000 md sand samples subjected to different overburden stresses. The following is a summary of the findings:The experimental data suggest that "fracture" propagation in unconsolidated sand is primarily a result of shear failure in a process-zone ahead of the fracture tip. The shear failure is caused by large tip stresses (due to tip plasticity) and by pore pressure increase within the process zone.Uncharacteristically large net fracturing pressures (NFP) were encountered for low-efficiency fluids. Generally, the NFP increased with decreasing fluid efficiency.Multiple sub-parallel fracturing and complex fracture geometry was encountered. sub-parallel fractures may be initiated at the tip or at the fracture wall due to shear failure and is dependent on the fluid efficiency and the type of leakoff, i.e., wall building or viscous. The field consequences of sub-parallel fracturing during stimulation may include pre-mature screenout and fracture bridging, short fracture length and extensive formation damage as the fracturing fluid invades the sheared interfaces. Introduction Fracpack completions are frequently performed in poorly consolidated, high-permeability sand formations, such as those in the Gulf of Mexico, to bypass well damage and enhance productivity. Generally, this stimulation involves injecting a small pad of clean fluid, followed by slurry with as much as 15 ppa proppant concentration, depending on the sand carrying capacity of the fluid. As much as 200,000 lbs of sand may be injected in one zone and fracture lengths of over 40 ft may be obtained. Many fracpack designs include tip screenout (TSO) to induce an inflated, highly conductive fracture that is packed by proppant as the fluid leaks off.1,2,3 Depending on the formation properties, a variety of fracturing fluids may be used in a fracpack operation, including borate cross-linked guar, linear gel, VES and non-viscous fluids.4–7 The cross-linked fluids, because of their wall building capability, can create relatively long fractures, but may cause extensive formation damage if the polymer breakdown is incomplete following stimulation. On the other side of the spectrum, non-viscous fluids result in minimal formation damage with the drawback that the resulting fracture is typically very short (less than 5 feet) due to excessive leakoff. These fluids as well as the linear gel and VES must be laden with fluid loss additives to induce effective fractures in high-permeability formations in excess of 500 mD.
A rate of penetration (ROP) prediction equation for an insert roller cone bit is determined from laboratory drilling tests as. a function of bit weight, well depth and laboratory measured rock properties. A complete description of each of the seven rock types used in the study is presented. A comparison of ROP at different flow rates for each rock shows minimal hydraulic cleaning problems at the base hydraulic energy levels chosen. The predictive equation will be valuable in predicting drilling rate and understanding how the rock properties affect it.
Summary A study was carried out to determine the geomechanical effects of polymer flooding in an unconsolidated-sand reservoir. The work involved laboratory-scale polymer injections in unconsolidated- sand blocks to identify the injectivity mechanisms, numerical analyses for fracture prediction, and geomechanical modeling of the formation to examine the potential of shear failure and containment loss during flooding. Laboratory tests under polyaxial conditions indicate that nearwellbore fracturing and permeability increase in unconsolidated sands occur at net injection pressures limited to 2.0 MPa. These findings were applied to fracture modeling. Geomechanical modeling suggests large-scale shear failure in the sand and in the bounding shale during polymer flooding. These are expected to affect both the fracture containment and the vertical-hole integrity. Finally, fracture predictions underscore the importance of the geomechanical considerations on determining the fracture dimensions and containment. Sensitivity analyses also point to the significance of binding several key parameters for fracture prediction. These include sand shale stress contrast, fluid quality and total-suspended-solids (TSS) content, fluid rheology and effective viscosity in the formation, and the filter-cake properties in the presence of polymer. This paper provides a geomechanical perspective on the generally complex problem of polymer flooding in unconsolidated formations containing viscous oil. The work also offers some insights into the critical issues that must be examined in such situations to avoid catastrophic failures. It highlights the existing technological gaps in the current predictive capabilities.
Summary Long-term stability of horizontal wellbore completions with uncemented liners in weakly consolidated to unconsolidated sandstone formations (e.g., the Gulf of Mexico and Nigeria) remains an area of concern. In this paper we present the results of dedicated polyaxial cell laboratory experiments that address this issue. In addition, the influence of rock failure in the near-wellbore region on well productivity was studied. Large blocks of a weak artificial sandstone were prepared. A hole was drilled in these blocks, and production conditions at various values of in-situ stress, drawdown and water cut, both in the absence and presence of a liner, were simulated. During testing, the hole was kept at a horizontal position in order to realistically simulate the influence of gravity forces on the movement of sand debris. The process of hole failure and re-stabilization was continuously monitored by an endoscope coupled to a video camera. The experimental results show that in the presence of a slotted liner, and in the absence of a water cut, rock failure leads to gradual annulus fillup with loose sand, eventually resulting in a stable configuration in which only a small fraction of the far-field stresses is transferred to the liner. These results are further supported by elasto-plastic calculations. Rock failure around the liner is shown to have only a minor effect on productivity. This result implies that rock failure around uncemented liner completions will generally not be noticed at the wellhead. The introduction of a small (<5%) water cut resulted in massive sand production and subsequent liner collapse. This can be explained by the fact that a water cut destroys capillary cohesion, thereby destabilizing sand arches over the slots. Introduction Reliable predictions of sand production potential are required to make realistic sand production management and contingency planning possible. Unnecessary application of sand exclusion measures results in increased completion costs and considerable loss of well productivity. Further, sand prediction may assist in selecting the most attractive sand control techniques. 1 Over the years, a large number of models for sand production prediction have been developed; see, for example, (Refs. 2-9). These models generally focused on the prediction of the onset of sand production. A new conceptual model for an initial sand production prediction is presented in another paper.10 However, in many situations a certain degree of sand failure around the wellbore, and resulting sand production, is acceptable within limits. In this paper we present laboratory experiments focusing on re-stabilization after initial sand failure around horizontal wellbores, with or without uncemented liners, in weakly consolidated to unconsolidated sandstones. Long-term stability and productivity of these completion types remains an area of concern in many fields, e.g., the Gulf of Mexico, Nigeria, North Sea, etc. Large blocks of a weakly consolidated (0.46 MPa cohesion) artificial sandstone were prepared, and a hole was drilled in these blocks. The effects of in-situ stress, flow rate (drawdown), water cut, and completion type (openhole, pre-drilled liner, slotted liner) on stability and productivity were investigated. During testing, the hole was kept at a horizontal position to realistically simulate the influence of gravity forces on the movement of sand debris. The process of hole failure and re-stabilization was continuously monitored by an endoscope coupled to a video camera. The liners used in the tests were made of (transparent) plexiglass in order to monitor the failure process around the liner with the endoscope. The experimental results show that, in the absence of a liner, increasing the far-field effective stress leads to gradual hole closure instead of a sharply defined "failure." In the presence of a slotted liner, and in the absence of a water cut, the endoscope images showed gradual annulus fillup by loose sand with increasing far-field stress. Finally, a stable configuration was achieved in which the liner was completely surrounded by loose, unproduced sand and plastically deformed sandstone, as was also confirmed by post-test computer tomography (CT) scans. This configuration was still stable at a vertical effective stress equal to three times the measured collapse strength of the liner. These observations were further supported by elasto-plastic calculations, which showed that redistribution of stresses around the hole resulted in forces on the liner which were only 5 to 25% those of the far-field effective stresses. However, the introduction of only a small (5%) water cut was sufficient to disturb the above-mentioned stable configuration, and resulted in massive sand production and subsequent liner collapse. This can be explained by the fact that the water cut destroys capillary cohesion, thereby destabilizing sand arches over the slots. Post-test CT scans showed large, horizontally oriented, washouts adjacent to the wellbore. The resulting stress concentrations on the top and bottom of the liner finally caused its collapse. The experiments showed that, in the absence of a water cut, rock failure around the liner only results in a small change in productivity. These results were further confirmed by elasto-plastic calculations. One implication is that no observed productivity decrease in the field does not mean that there is no rock failure, which may have consequences, e.g., for future water breakthrough (sand production) and for selective placement in stimulation/shutoff. In the next section we provide a description of the laboratory setup and the experiments that were conducted. Then we present and interpret the experimental results. In the last section conclusions are drawn. Experimental Approach A total of six polyaxial cell tests were performed on low-strength artificial sandstone blocks (26.25 cm×26.25 cm×38 cm) with 25.4 mm diameter horizontal holes. Artificial sandstone was used rather than a natural poorly consolidated rock (such as Saltwash South) since natural low-strength rocks have frequently been observed to have highly variable mechanical properties and physical characteristics. An artificial rock with consistent properties was fabricated using a stringently controlled manufacturing program.
Long-term stability of horizontal wellbore completions with uncemented liners in weakly consolidated to unconsolidated sandstone formations (e.g. Gulf of Mexico, Nigeria) remains an area of concern. This paper presents the results of dedicated polyaxial cell laboratory experiments addressing this issue. In addition, the influence of rock failure in the near-wellbore region on well productivity was studied. Large blocks of a weak artificial sandstone were prepared. A hole was drilled in these blocks, and production conditions at various values of in-situ stress, drawdown and watercut both in the absence and presence of a liner, were simulated. During testing, the hole was kept at a horizontal position in order to realistically simulate the influence of gravity forces on the movement of sand debris. The process of hole failure and restabilisation was continuously monitored by an endoscope coupled to a videocamera. The experimental results show that in the presence of a slotted liner, and in the absence of watercut, rock failure leads to a gradual annulus fill-up with loose sand, eventually resulting in a stable configuration in which only a small fraction of the farfield stresses is transferred to the liner. These results are further supported by elasto-plastic calculations. Rock failure around the liner is shown to have only a minor effect on productivity. This result implies that rock failure around uncemented liner completions will generally not be noticed at the wellhead. The introduction of a small (<5%) watercut resulted in massive sand production and subsequent liner collapse. This can be explained by the fact that watercut destroys capillary cohesion, thereby destabilising sand arches over the slots. Introduction Reliable predictions of sand production potential are required to make realistic sand production management and contingency planning possible. Unnecessary application of sand exclusion measures results in increased completion costs and considerable loss of well productivities. Further, sand prediction may assist in selecting the most attractive sand control techniques. Over the years, a large number of models for sand production prediction have been developed, see e.g. Refs. 2-9. These models generally focussed on the prediction of the onset of sand production. A new conceptual model for initial sand production prediction is presented in an accompanying paper. However, in many situations a certain degree of sand failure around the wellbore, and resulting sand production, is acceptable within limits. This paper presents laboratory experiments focusing on re-stabilisation after initial sand failure around horizontal wellbores with or without uncemented liners, in weakly consolidated to unconsolidated sandstones. Long-term stability and productivity of these completion types remains an area of concern in many fields, e.g. Gulf of Mexico, Nigeria, North Sea, etc. Large blocks of a weakly consolidated (0.46 MPa cohesion) artificial sandstone were prepared, and a hole was drilled in these blocks. The effects of in-situ stress, flow rate (drawdown), watercut, and completion type (open hole, predrilled liner, slotted liner) on stability and productivity were investigated. During testing, the hole was kept at a horizontal position in order to realistically simulate the influence of gravity forces on the movement of sand debris. The process of hole failure and re-stabilisation was continuously monitored by an endoscope coupled to a videocamera. The liners used in the tests were made out of (transparent) plexiglass in order to monitor the failure process around the liner with the endoscope. The experimental results show that in the absence of a liner, increasing the far-field effective stress leads to gradual hole closure instead of a sharply defined 'failure'. In the presence of a slotted liner, and in the absence of watercut the endoscope images showed a gradual annulus fill-up by loose sand with increasing far-field stress. P. 35
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.