Abstract

As the inclination of a wellbore increases, cuttings start having a tendency to accumulate in the lower section of the wellbore, and develop a cuttings bed. This developed bed causes a reduction in the flow area, an increase in the friction between the drillstring and the wellbore, leading to an increase in torque, decrease in force transfer to the bit, and lose control on the bottomhole pressure. Finally, drilling rate decreases. Estimation of total concentration inside the wellbore is not an easy task. Moreover, moving cuttings and fluid dragging them to move have different relative velocities inside the wellbore, causing variations in pressure drop. Although there are many attempts to estimate the cuttings concentration and slip between the cuttings and the fluid using mechanistic models and empirical correlations, the performances are limited either with the strong assumptions made, or the experimental facility capabilities. This study aims to determine some of the very-difficult-to- identify data for estimating total pressure drop and total cuttings concentration inside the wellbore. Extensive experimental work has been conducted using a cuttings flow loop at horizontal and inclined wellbores. Tests have been conducted using water to simulate low viscosity fluids. Data has been collected for a wide range of flow rates, cuttings injection rates and pipe rotation speeds. All experiments have been recorded using a high-speed digital camera. Images have been processed using special algorithms, and volumetric distributions of cuttings and fluid can be identified very accurately. By comparing consecutive images, very valuable information has been collected about the accumulated cuttings amount, concentration of moving particles, their relative transport velocities, slip velocity between the phases, the friction factor on the stationary bed, etc. Since the images are digital, information collected is converted into numerical values, and semi-empirical equations are developed as a function of known drilling parameters. The obtained information is tested in simple mechanistic models for estimating pressure drop inside a wellbore with the presence of cuttings, and the performance of the model is tested by comparing the results with the measured ones. It is observed that, after supplying the very-difficult-to-identify information to the mechanistic model, the performance of the mechanistic model improved very significantly. The information provided will improve the design of long extended reach wells while estimating hydraulic requirements, and make it possible to have a better understanding of what is really happening inside the wellbore.

You can access this article if you purchase or spend a download.