Optimization of Horizontal Well Completion Design

Abstract

Abstract A well completion is a critical interface between the productive formation and the wellbore. An effective completion must maintain mechanical integrity of the borehole without creating any significant restrictions in the flow capacity of the well. In this paper, we outline a process to design optimal completions for horizontal wells by applying comprehensive skin factor models that include damage and turbulence effects for all common types of completions. Slotted or perforated liner, cased and perforated completions, or gravel pack completions have been used in horizontal wells for borehole stability and sand control purposes. However, these completions may have lower productivity (as characterized by a positive skin) relative to an equivalent openhole completion because the convergent flow to perforations or slots increases fluid velocity in the near-well vicinity. In addition, any reduced permeability zones (formation damage caused by drilling, completion, or other processes) magnify the convergent flow effects, and hence, may result in severe skin factors. Compound effects of formation damage around the well completion, a crushed zone due to perforating, the plugging of slots, and turbulent flow, as well as interactions among these effects are included in the model. We first illustrate how to use the skin factor models to screen the available completion types for different applications. This screening approach considers reservoir permeability, permeability anisotropy, fluid properties, formation damage effects, and rock mechanical characteristics as the key parameters. The types of completions that yield the most productive well performance for this matrix of properties are presented. A more detailed completion design is then illustrated by showing the use of the skin factor models for selection of liner completions for viscous oil reservoirs on the North Slope of Alaska. Application of the slotted or perforated liner models to the readily available liners showed that the completion skin factor can vary by as much as 40%, depending on the detailed characteristics of the slots or perforations in the liner (slot or perforation size, density, and distribution). This showed how analyzing the performance of the completion design can greatly increase well productivity at little or no cost. Introduction The optimization of well completions to improve the inflow performance of horizontal wells is a complex but very practical and challenging problem. What is needed is a means for engineers to determine the causes of high skin values occurring under various conditions and suggest techniques to minimize the problems. In particular, the interactions among formation and perforation damage effects and convergent flow to perforations and slots are critical issues to design optimal completions for horizontal wells. Numerous papers have reported on completion performance models of vertical and horizontal wells, which can be used to predict the productivity of the wells. These models can be categorized into two groups; the numerical models1–3 and the (semi-) analytical models4–6. The numerical models require advanced computer programs to solve the complex flow problem by applying the finite difference method, the finite element method, or the Green's function (source function) method. All the methods require the solution of a large matrix system and relatively higher computational time compared with analytical models although they provide accurate solutions for various conditions. Analytical and semi-analytical models are widely accepted in field practices because they are usually easy to use and provide a better understanding of the relative role of various parameters in affecting well productivity. These models are obtained by solving the complex flow problem with simplified assumptions and are sometimes calibrated by numerical simulation results. Furui et al.7 presented a mathematically rigorous general form of a skin equation, which consists of the rate-independent skin factor, s0, the turbulence scale factor, ft, and the Forchheimer number, Fo (note that the product, ft·Fo is equivalent to the rate-dependent skin factor). According to their model, the skin factor for any type of well completions is expressed in the form of:Equation 1 where the subscript, i denotes the type of well completion.