Abstract

Two-phase flow models play a fundamental role in the design, improvement, and operation of oil and gas production systems. Considering very large high-viscosity oil reserves around the world, the understanding of two-phase flow systems, which involves high-viscosity-liquids (>100 mPa s), becomes a relevant topic. The behavior of this kind of flows is expected to be significantly different as compared with low viscosity oils. In this work, we report a new experimental data on high viscosity oil-gas flow. A set of 170 upward flow experiments were carried out using a mixture of air and mineral oil (213 mPa s, 50.8 mm ID, 21.5 m long) for five different elevations: 45°, 60°, 70°, 80°, and 85°. Superficial liquid and gas velocities were varied from 0.05 m/s to 0.7 m/s and from 0.5 m/s to 6 m/s, respectively. We measured and analyzed: flow pattern, pressure gradient, and average liquid holdup. These variables helped evaluate (a) the implications of a buoyancy-like term in the mechanical energy balance, and (b) the performance of two-phase flow models under the experimental conditions. As a result, we propose a practical relationship between the actual and homogenous densities through the slip and no-slip holdup fractions. This approach, supported by experimental evidence, shows that considering a buoyancy-like term is congruent with the physics that results from the interaction of gravitational and friction forces in inclined pipes. Moreover, the comparison of the present data with prediction models shows a performance which varies significantly depending on the predicted variable, the existing flow pattern as well as the inclination angle. In general, the model performance is good when the observed flow pattern is well predicted but fails when it is not, which typically falls into churn flow. This justifies the need to further understand churn flow and develop models consistent with the physics of transport.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call