| dc.description.abstract | To provide zonal isolation and avoid leakages behind the casing it is important to obtain a high quality cementing job. High quality cement is a result of a good drilling fluid displacement. In order to be able to control the cement flow dynamics in a realistic well scenario, it is important to predict the flow pattern in the annulus and thus the position and shape of the cement interface. In a concentric annulus, it can be assumed that the distribution is radial homogenous in the annulus. But in a realistic scenario today, where it is becoming more common to drill complex and deviated wells, the casing string will often be concentric positioned in the well, causing an eccentric annulus, i.e. where the casing string is not 100 % in the centre of the borehole.
Present project has investigated flow distribution in eccentric annuli. There is limited amount of mathematical models describing this distribution, and thus it has been focused on possible methods to obtain empirical data on this. Some companies such as ConocoPhillips and Halliburton have carried out research on the topic and conducted experiments on flow distribution in eccentric annulus. However, it is still required to find better low-cost methods of measuring the fluid velocity distribution. Some possible methods have been considered and discussed in this project, and the plan of conducting an experiment based on one of these methods is described in detail.
This method is based on Bernoulli equation and stagnation pressure. By measuring the dynamical and static pressure at different angular positions around the annulus, one can calculate the local fluid velocity at the measured point. This experiment requires equipment such as an eccentric annular model, a pump, small pivot tubes to register dynamic and static pressure, and a suitable pressure transmitter. Some of this equipment has, however, shown to be more intricate then expected to get hold of in time within the project timeframe. This led to difficulties in performing the actual experiment within the time extent of this project. A description of how the experiment was planned and analyzed is being described, and a suggestion of how to accomplish this is being presented in a Gantt-Diagram.
A mathematical model of Newtonian flow velocity profile distribution was made in Excel and Matlab for the purpose of comparison to real test results and how to make this simulation more realistic is discussed.
The main findings of this work is that to satisfy conditions discussed in the project, the most feasible method for measuring the flow velocity in eccentric annulus is using pressure measurements. A plan for performing this experiment is developed, and all the experience gained during this process has been used as a basis for improvement and further follow-up. | nb_NO |
