Abstract There has been substantial recent progress in coupling geomechanical effects to reservoir response, thus dramatically improving representation of the consequences of rock response to pressure changes. New reserves have been identified from compaction. Injection geomechanics has gained an increase in interest due to its impact on the reservoir, faults and well hardware. Changes in transmissibility are now seriously implemented in reservoir engineering tools with more attention directed at the presence of compaction and dilatant bands around producers or injectors. Efforts are progressing to ensure adequate coupling between the local effects of the rock deformation near the wellbore, as well as along faults or bedding planes, and the evolving stresses and deformation in the reservoir. This paper attempts to discuss current gaps in understanding the intricacies and details of coupling farfield deformation to well completion. Examples are shown where reservoir compaction or dilatancy is explicitly coupled to near-wellbore behavior, with specific application for assessing well performance and survivability. The analyses can use reservoir simulations coupled with analytical predictions of stresses and deformations in individual simulator blocks. The predicted stresses and deformations form the boundary conditions for finite element modeling that can focus in on the details around the completion itself. This is in contrast to the current approaches that use explicit coupling of pressure and deformation in complete massive finite element representations, with refined gridding around the completion. The intimate details of coupling reservoir deformation to the completion require more intensive consideration. For example, "How can the cement sheath be represented?" or "What are some of the constitutive considerations in the near-wellbore region that impact integrity?" and "How is varying transmissibility related to well integrity?" These issues are considered. There are three goals: Start to recognize completion and production management practices that will improve completion longevity and optimize well productivity. Identify reasonable methodologies for representing the coupling between the completion and the reservoir, including yielded zones (dilatant and/or compactant), compaction bands with varying transmissibility, the cement sheath with or without a microannulus and the mud cake. Delineate approximate methods that will adequately forecast completion distress and permeability impairment without the necessity of expensive and time-consuming detailed finite element simulations. Introduction The industry has been experiencing a surge in activities related to the exploitation of reservoirs under complex conditions such as: Deepwater, under-saturated, abnormally pressured and unconsolidated sands where compaction drive can lead to subsidence and casing deformation in costly wells, especially subsea wells, and where future interventions are prohibitive and time consuming. Sand-production prone reservoir layers where sand exclusion imposes completion requirements that may impede achieving maximum well productivity. Depleted sand and carbonate zones where loss of circulation during infill drilling is undesirable and the presence of natural fractures may aggravate the situation. Horizontal, high angle and extended reach wells where well integrity during the well life is necessary. Environmental requirements that preclude emission of the associated produced streams (gas, water, drilling and completion wastes, etc.) and hence, may require provision for long-term subsurface injection of slurries.

Abstract The Valhall field, located in the Norwegian Sector of the North Sea, is a high porosity chalk field characterized by its weak reservoir rock. Global field behavior during depletion includes extensive compaction and associated subsidence. One highly successful completion alternative, horizontal wells, was, in its early application to Valhall, plagued by the premature failure of the production tubulars associated with reservoir draw down/compaction. Numerical modeling, with subsequent field implementation, has shown, however, that casing with sufficiently low diameter:thicknesses can withstand the rigors of draw down and associated formation loads. Unfortunately, the use of thick-walled casing presents its own difficulties in terms of installing the string (torque/drag) and the subsequent limitations on tool diameter (smart well completions, for example) associated with the stout cross section. Further, a one-size-fits-all doctrine is not optimal, in that the porosity, and therefore strength, of the reservoir rock varies laterally, thus penalizing well designs on the flanks of the reservoir where the chalk porosity is lower and chalk strength is, therefore, higher. The current paper addresses the above issues by modeling the behavior of the wellbore and surrounding region. Starting with the virgin reservoir, the model considers (sometimes extensive) pre-wellbore depletion, drilling the wellbore, installation of casing and cement, and subsequent draw down. Significant variations to the model include the cross section of the production casing, the quality and quantity of the cement sheath and the perforation pattern. Behavior of these various configurations during subsequent draw down permits them to be ranked according to the life expectancy of the resulting completion. The discussion is enhanced by field results, as several of the completion alternatives have been installed in Valhall wellbores. Introduction In an attempt to increase both the quantity and rate of recovery of hydrocarbons from the Valhall field, horizontal wellbores were introduced on March 22, 1991, with the completion of Well 2/8-A-12BSt1. Since that time a number of horizontal wellbores have been drilled, almost exclusively on the flanks of the field initially, but later also in the crest of the field. Not surprisingly, a number of the initial horizontal completions, including cemented and uncemented liners and concentric liner configurations, failed or otherwise proved unsatisfactory. The Valhall producing environment is severe, the formation chalk undergoing extreme compaction during depletion 1–3 , with associated consequences for wellbores and tubulars penetrating the reservoir. At the other extreme, installation of unduly thick tubulars can result in difficulties both in running the string (e.g. torque, drag) and with regard to clearance restrictions in the implementation of smart completions. The current study addresses the integrity of Valhall horizontal wellbores via a number of two-dimensional finite element simulations intended to model the entire history of the reservoir from discovery, through global pore pressure depletion, to the introduction of a wellbore, and, finally, through local production draw down. Accurate modeling is hampered by the extreme deformation that may occur prior to the introduction of the wellbore and by the uncertainty of cement coverage over long (1–2000 meter) horizontal sections. The discussion begins with a description of the analysis variables, including reservoir and borehole conditions and the character of the casing, cement and formation chalk. A detailed description of the numerical modeling methodology is included to aid readers who may want to duplicate this analysis. These preliminary topics are followed by results for a number of completion, depletion and cementing scenarios.

Abstract This is a case study of a successful, field-wide implementation of horizontal, barefoot completions in a moderately competent formation - the Alpine Reservoir in Western North Slope, Alaska, Figure 1. A barefoot completion is a borehole without tubular liners and no cemented support - the least costly but riskiest completion strategy. (In this paper, barefoot and unsupported mean the same and are used interchangeably.) The Alpine experience provides a successful example where the benefits of unsupported boreholes outweigh the risks of borehole failure. To date, this aggressive yet simple completion technique has an aggregate length of more than 160,000 ft, all unsupported. Combined with good drilling practices, the success of barefoot horizontal wells in Alpine is also due to the following petrophysical and geomechanical factors: Consistent reservoir quality within the layer Absence of shale-breaks in the producing zone Moderate strength in normal fault geotectonic setting Low variability of strength Linearly elastic behavior Non-severe, slight weakening when water-saturated Good permeability retained under post-elastic strains Introduction In general, barefoot horizontal completions are implemented only in very competent, hard formations that pose little risk for wellbore collapse and/or sand production, such as dolomites, hard limestones, hard sandstones, and shale-free siltsones. The advantages of barefoot completions are: Low completion cost Simple and fast implementation Potentially lower completion skins if undamaged High productivity per unit length (producers) High injectivity (injection wells) Higher critical drawdown pressure for sanding However, the risks and disadvantages of unsupported, long, horizontal wells in Alpine are: Potential for collapse or sanding in weak zones Higher wellbore stresses compared to vertical wells Costly and limited options for zonal isolation Higher sensitivity to formation damage Limited options for stimulation Limited options for future remediation High friction factors for future coiled tubing workover There are very few reservoirs that are developed exclusively with barefoot horizontal wells, owing to the abovementioned risks. One recent example is described in Australia (Allard, 1998). The Alpine experience is the first reservoir in Alaska developed exclusively with barefoot horizontal wells.