Eulerian Model Performance in CFD Calculating the Drag Coefficient from Drift-Flux Model for Gas-liquid Two-phase Flow in Vertical Pipes

Printer-friendly versionPDF version

The processes efficiency and profitability can increase by the two-phase flow characterization. For example, the petroleum industries must know the natural gas and oil quantities coming from a reservoir to calculate the production profitability. However, designing tools for this purpose is an engineering challenge. Experience shows that computational tools are more efficient than experimental tools. Currently, the Computational Fluid Dynamics (CFD) is the most used tool for these assessments. Commonly the two-phase flow modeling in this technique is made by the Volume of Fluid model (VOF) or the Eulerian model. The mathematical formulation of both models causes a difference in their convergence, speed and accuracy. The VOF model does not distinguish the velocities between phases and the Eulerian model can adjust with other models as the drift-flux model to increase its accuracy. In this project, six experimental studies were modeled with CFD. The results show that Eulerian model has a lower root mean square error (18.25%) than VOF model (22.93%). Furthermore, the results demonstrate that the Eulerian model accuracy is independent from the grid, demanding less computational time than the VOF model. Finally, it was verified that adjust the Eulerian model with the drift-flux model causes convergence problems. 

Author Company: 
Universidad de los Andes
Author Name: 
Hugo Pineda