Optimization of Displacement Efficiency During Primary Cementing

Abstract

Different philosophies exist for how to design cementing jobs in order to achieve the best possible displacement. This is due to the fact that the process is complex and that it often is difficult to measure the degree of displacement in a well. The trend towards extended reach wells and slim hole wells gives additional challenges when trying to achieve good displacement. A large amount of high quality displacement experiments have been performed in a 4m long laboratory scale loop. Laminar and turbulent flow were considered and densities, rheologies, inclination and eccentricity were varied during the experiments. In all the experiments the displacement efficiency was accurately measured. The experimental displacement data has been analyzed with statistical methods. The results from these analysis and additional simulation results have been used to develop a correlation based displacement model. This model is based on some physics since relevant non dimensional parameters are used in the correlation. This correlation gives a fast prediction of the displacement efficiency and is an efficient engineering tool for optimizing the displacement efficiency. Modeling of the displacement process is complicated due to fully three dimensional geometry and flow of non-Newtonian fluids with different rheologies and densities. The displacement process has been modeled and simulated using a general three dimensional flow program. Results from the simulations have been compared against the experimental results. Results from simulations are found to compare well for the cases considered. The general flow simulator is based on basic physical laws. This enables accurate prediction of displacement efficiency at field conditions. Introduction Investigators have performed laboratory experiments in order to study the displacement process in detail. Tehrani et. al. did experiments for laminar conditions in an annulus formed by two pipes. They found that when covering a wide range of parameter space. the complexity of the displacement process did not allow for maIling straight forward global rules for improving the displacement efficiency. Silva et. al. performed displacement experiments in an annulus formed by an inner pipe and an artificial permeable formation. A complete cementing operation was performed including pumping of a drilling fluid with cuttings, spacer, washer and cement. The cement was allowed to set and the section was cut in slices for visual inspection. They found that the axial velocity component as a function of the radial distance in the annulus should be as flat as possible in order to achieve the best displacement. Lockyear et. al. and Ryan et. al. observed that a necessary condition for efficient displacement is that the pressure gradient has to exceed the combined forces due to drilling fluid yield stress and the density difference between cement and drilling fluid. Attempts have been made to use analytical methods to model the displacement process. Flumerfelt investigated laminar displacement in a vertical concentric annulus within the parallell plate approximation. Nguyen et. al. studied displacement in a concentric horizontal annulus. These investigations are of limited value because of the assumptions done in order to obtain analytic solutions. Haut et. al. and Graves et. al. computed displacement of one non-Newtonian fluid by another with higher density in a vertical, concentric annulus. Results from the simulations illustrated the temporal evolution of vortices and instabilities. Szabo and Hassager studied the displacement of one Newtonian fluid by another with higher density in a vertical annular geometry. Both analytical and numerical methods were used to study the displacement process.

Primary cementing of horizontal wells. Displacement flows in eccentric horizontal annuli. Part 1. Experiments

“…by rotation, and this has been addressed by Carrasco-Teja & Frigaard (2009), Carrasco-Teja & Frigaard (2010), Tardy & Bittleston (2016) and Tardy (2018). Three-dimensional simulations of annular displacement flows were first developed in the late 1990s but were too slow for practical usage (Szabo & Hassager 1997;Vefring et al 1997). In recent years interest has revived in these methods as open source and other computational codes have become widely accessible and multi-processor computations have made such calculations manageable at reasonable resolutions over increasingly long annuli and even for non-Newtonian fluids, e.g.…”

An experimental study of laminar displacement flows in narrow vertical eccentric annuli

“…Other than the studies by Tehrani and coworkers, there have been only occasional experimental results reported (e.g. Jakobsen et al 1991;Vefring et al 1997;Dutra et al 2004;Nguyen et al 2008).…”