Summary. Thaw subsidence during the production life of wells drilled through the permafrost can transfer very high axial compressive loads to the surface casing. Improved techniques were required to define the compressive strain capacity of surface casing for North Slope operations. These techniques were used to identify cost-effective surface-casing designs capable of withstanding the axial compressive strains imposed by permafrost thaw. Plans to drill North Slope development wells on closer surface spacing resulted in predicted permafrost thaw loads on the surface casing that exceed current operating practices. In addition, development of lower-productivity wells has resulted in completions with smaller-diameter tubulars. Both of these considerations have led to a reassessment of the surface-casing programs for these new wells. The results presented are based on two parallel and complementary investigations: full-scale physical testing and nonlinear finite- clement analysis. The physical tests provided information about the performance of the casing and connections under high compressive loads. Data were used to verify the accuracy of corresponding finite-element models with three different criteria: strain-gauge data from the coupling and pin surfaces, failure mode comparison, and failure load comparison. An excellent match was achieved between the strain-gauge data and the analytical strain results, the actual vs. modeled failure mode (i.e., compressive bulging in the pipe body), and the actual vs. modeled failure load. Once verified, the models were used to investigate the effects of field-related parameters that were not tested. These included variations in makeup condition, pipe-wall thickness, thread taper angle, and casing length. Finally, these results enabled the selection of suitable surface-casing materials and connections for arctic operations. Introduction Surface casing is set through the permafrost for Alaskan North Slope wells (as deep as 2,000 ft [610 m] in some areas) to establish an adequate "foundation" for the wellhead and tree assembly, to provide backup pressure containment, and to isolate the inner strings from permafrost loads. Permafrost thawing caused by the long-term production of high-rate oil wells can lead to pore-pressure reduction, resulting in sediment compaction. The magnitude of strain that these effects induce on the surface casing depends on several factors, including the mechanical properties and lithology of the permafrost and the proximity of nearby wellbores that contribute to the overall thaw area. The incentives to reduce the wellhead spacings of arctic wells include limiting gravel requirements for both offshore islands and onshore well pads on the tundra and reducing the length of pipe- line required to manifold the wells. Proposed wellhead spacings (as close as 15 ft [4.6 m] for offshore islands) were expected to produce higher soil strains than those predicted for original Prudhoe Bay wells spaced roughly 120 ft [37 m] apart. Therefore, a full performance study was conducted to quantify the axial strain limits of 13 3/8-in. [34-cm] -diameter, 72-lbm/ft [107-kg/m], L-80 butress-threaded casing (common surface casing in North Slope use). The performance was compared with that predicted for normalized N-80 grade casing used on some early Prudhoe Bay wells. Description of Analysis Methods The casing performance was analyzed with the finite-element method and verified by well-instrumented, full-scale testing. Non- linear finite-element analysis techniques have been developed for the analysis of threaded connections with a development version of the ABAQUS general-purpose finite-element program. Additionally, processing hardware and software were used to allow efficient analysis of large finite-element models on the Cray X/MP supercomputer. This processing and analysis system allows efficient and accurate creation of large finite-element models and aids greatly in the interpretation of the results. Finite-Element-Model Generation. Three initial assumptions used in creating the models wereaxisymmetry,symmetry about the plane through the center of the coupling, anddisplacement-controlled axial loading at the ends only. The two symmetry assumptions greatly reduce the size of the model and the associated computing cost. The third assumption allows no shear loading on the outside of the casing, only displacement-controlled end loading. This loading constraint corresponded to the physical test con-ditions. Permafrost thaw subsidence causes displacement-controlle loading on the casing in the field, although the loading is applied by shear forces on the outside of the casing. Several finite-element meshes were created and tested during the course of the casing analysis. First, test meshes were developed for the buttress threadform itself. Next, test meshes were developed for the pipe and coupling walls. For a given model, or model subset, a finer mesh with smaller elements will give a more accurate solution. Computer costs associated with a given model increase approximately with the square of the number of elements in the model. The final finite-element mesh selection is a tradeoff between cost and accuracy that was resolved by examining strain jump and radial stress contour plots and considering the need to retain fidelity in the threadform geometry. A typical finite-element mesh for the buttress connection is shown in Fig. 1. This mesh contains about 4,500 nodes and 2,500 quadrilateral constitutive elements (both second-order, reduced-integration and first-order, full-integration). Nonlinear Analysis Techniques. Three major areas of nonlinear behavior must be modeled accurately to analyze the performance of the casing with buttress-threaded connections:material properties,deformation geometry, andsurface interaction between the pipe and coupling in the thread region. These nonlinearities must be handled with an iterative solution procedure. For a solution to satisfy the convergence criterion. equilibrium must be satisfied within a specified tolerance at every degree of freedom. To model the history dependence of the final solution accurately, equilibrium convergence must be achieved for each incremental solution during loading. This process required more than 500 iterations for each model. SPEDE P. 289^
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.