Applications Of A Computer Simulator To Primary Cementing

Abstract

Abstract An investigation of the relative importance of the various physical and rheological properties of the fluids involved in a primary cementing operation has been carried out using a computer simulation of the displacement process. The difference in densities between drilling fluid and cement has been found to be a major factor controlling displacement efficiency and the formation of the interface. Guidelines governing optimum density differences are presented. Introduction The effective displacement of the drilling fluid by the cement slurry and the mixing of the two fluids during primary cementing operations are critical factors in the successful completion of oil and gas wells. Recently, Graves and Collins, working under contract with Exxon Production Research Co., developed a computer simulator that provides an accurate solution of the complete set of hydrodynamic equations governing the displacement of one non-Newtonian fluid by smother in an axially symmetric system. Their computer simulator has been used successfully to study the primary cementing process. A systematic study of the relative importance of drilling fluid and cement rheologies and densities as they affect displacement efficiency and the interfacial profile has shown that the density difference between the drilling fluid and the cement is a major factor controlling the displacement efficiency and the formation of the interface. PRIOR INVESTIGATIONS PRIOR INVESTIGATIONS Several research projects (for example, Refs. 2 through 6) have been performed in the area of primary cementing. Although all of these research endeavors pertain to the displacement process, only a few have pertain to the displacement process, only a few have attempted to describe mathematically the cement-mud interface as it develops during the displacement of one fluid by another. Ref. 5 and 6 describe two different numerical studies that simulate the displacement process. Both of these studies made simplifying assumptions that were not made in this study. For example, the study discussed in Ref. neglects any density difference between the two fluids, assumes the displacement interface between the drilling fluid and the cement is distinct and horizontal, and neglects any radial velocity component. Although the study discussed in Ref. models the difference in densities between the two fluids, assumptions are made that the leading edge of the displacing fluid is well defined and stable, and that the flow is one-dimensional, having only an axial velocity component. Because of these simplifying assumptions, especially that of one dimensional flow, neither of these studies provided as realistic a simulation of the interfacial mixing that occurs during the displacement process as does this study. The nature of this mixing is described in the following sections. THE VELOCITY PROFILE As shown in Fig. 1, there are three Possible flow regimes in which a non-Newtonian fluid, such as a drilling fluid or a cement slurry, may exist. In this figure, the dashed lines represent the average axial velocity and the solid lines represent the actual velocity profile in the annulus. The axial velocity in the laminar flow regime is not as uniform across the annulus as it is in the plug flow or turbulent flow regimes. Because of the velocity distribution, the fluid near the center of the annulus will have a higher axial velocity component than the fluid near the boundaries. Velocity profiles for single fluids are very misleading if an attempt is made to use these to visualize the displacement of one fluid by another. The laminar velocity profile would suggest that the displacing fluid could "channel" through the center of the displaced fluid when the displacing fluid is in laminer flow. However, because of several instabilities that may develop, this does not always occur.