An integrated flow model is developed to simulate the flow into horizontal wells with screens. The model is composed of a multisegment horizontal-well model and a near-wellbore model. The multisegment horizontal-well model predicts the pressure losses in a bounded reservoir. The near-wellbore model simulates the fluid flow through the damaged/collapsed zone around the wellbore, the mudcake, the wellbore/screen interface, the gravel-packed annulus, and the screen itself.
Using the model, we investigated the effects of degree of screen damage, nonuniform damage removal, mudcake properties, and blank-pipe insertion on horizontal-well productivity. We examined and compared the long-term horizontal-well performance in terms of rate decline and cumulative production as a function of time.
Horizontal wells with sand-control problems may be completed with screens. Screens may be used in a variety of ways; prepacked screens, standalone all-metal screen-only completions, screens with openhole gravel packing, and premium screens with blank-pipe sections.
Prepacked screens may be preferred because of their lower cost. Historically, prepacked screens were used when there was a good chance of gravel-pack failure caused by incomplete packing. Prepacked screens may effectively minimize sand production. However, they have a smaller inflow area, which may cause productivity decline. Also, prepacked screens are vulnerable to plugging. Preventing the damage to prepacked screens may be the key to maximizing productivity. Screens may be partially or completely plugged. Localized partial plugging creates higher flow velocities at nonplugged screen sections, which in turn may initiate screen erosion. Several experimental studies have been dedicated to characterizing the damage to the screen. In several field applications in which the prepacked screens were the preferred completion method, sharp declines in well productivity were reported.
Horizontal open holes also may be completed with premium-quality, all-metal screens in standalone mode. Recently, a number of industrial studies focused on evaluating the performance of screen-only completions. When used in a standalone mode, the screens pro viding sand retention and resistance to plugging should be chosen.
If the annulus between the horizontal open hole and screen is gravel packed, then the amount of formation-sand production may be reduced significantly. However, gravel packing is more complex and expensive. Packing efficiency and the transportation of the gravel to the formation face could be the major issues. Incomplete packing in the annulus may lead to localized high-velocity hot spots and screen erosion.
When the premium-quality screens are used with or without gravel packing, the completion cost is high. If blank pipes are inserted between the expensive screen segments, then the completion cost may be reduced to a level that is comparable with the cost of prepacked screens. However, when the blank-pipe segments are distributed along the horizontal well, the contact area between the wellbore and reservoir is reduced, and well productivity is decreased.
There have been many studies focusing on the technical issues related to horizontal openhole completions with screens, including experimental studies, field applications, and well-productivity modeling.1–23 Experimental studies concentrated mostly on how to choose the screens that would perform best in a specific field.1–10 Guidelines for screen selection, screen performance, laboratory procedures, and the design of drilling and completion fluids can be found in the cited references. Refs. 1 through 5 describe the types, relative merits, sand-retention capability, collapse and erosion resistance, and plugging tendency of prepacked screens. Refs. 6 through 10 report the development and application of laboratory procedures for the evaluation of screen performance.
Details of several field case histories about cased-hole and openhole gravel packs are reported in Refs. 11 through 15. The field applications covered the East Wilmington field,11 the Gulf of Mexico fields,12 the Champion field, 13 and the Alba field.14,15
Inflow-performance models for horizontal open holes are relatively simple and well known.16–19 However, these models consider that the total drilled horizontal length is open to flow. In many cases, horizontal wells may be completed at selected intervals along the well axis. The horizontal wells partially completed at multiple producing segments are referred to as selectively completed horizontal wells (SCHW). Several modeling studies on the performance of the SCHW have appeared in the literature. 20,21. Although these models account for selective completion effect, they ignore the details of flow convergence caused by well completion.
There is a gap between the experimental studies/field applications1–15 and the well-productivity models. 16–21 Engineers should benefit from the results and methods suggested from both groups of work. Burton and Hodge22 and McMullan and Larson23 attempted to integrate the experimental and practical approach with the well-performance models. Burton and Hodge22 argued that screen plugging could severely impair well productivity if prepacked screen permeabilities were allowed to drop to approximately 10% of reservoir permeability. Recently, McMullan and Larson23 used a black-oil simulator to investigate the impact of blank-pipe insertions on the oil recovery from a horizontal well completed with premium screens.
To construct a simpler and faster model, the 3D flow into a horizontal well is decomposed to a transient SCHW with a general variable skin factor across the open intervals and a sectionally heterogeneous 1D radial near-wellbore flow model. The SCHW model predicts the pressure drop in the reservoir. The near-wellbore flow model accounts for fluid flow in damaged formation, across the mudcake, across the gravel pack, and through the screen. This approach significantly reduces the CPU time.
A multisegment horizontal well in a rectangular parallelepiped reservoir with impermeable external boundaries is considered. Multisegmentation allows us to account for the local changes around the wellbore. The SCHW assumes that the completed intervals are fully open to flow all around the perimeter of the segment. A variable local skin around each segment is also incorporated into the SCHW model. The additional pressure change caused by well completion is superimposed on the SCHW in terms of local skin. A schematic of a flow model for a multisegment horizontal well is given in Figs. 1 and 2.
The mathematical treatment of the SCHW model is presented in Ref. 24. The reader is referred to the cited reference for additional details.