A correlation for sand erosion prediction in annular flow considering the effect of liquid dynamic viscosity

Sand erosion is a common problem in oil and gas industry, which may lead to unpredictable failure and economic loss of pipeline systems. Due to the relatively stable state, annular flow is the predominant flow pattern and can simultaneously transport two kinds of materials (such as liquid and gas), which can reduce expenses for oil and gas production. In the present work, an accurate combination of the correlations of entrainment fraction and liquid film thickness is introduced into a modified numerical model, and is validated by experimental data. A series of numerical simulations are designed and performed to investigate the coupling effects of different parameters on erosion, such as the liquid dynamic viscosity, pipe diameter, superficial gas velocity, radius of curvature, superficial liquid velocity and particle size. Based on the knowledge gained from numerical analysis, the dimensionless pi groups are proposed as well as a correlation for predicting sand erosion in annular flow, which takes the effect of liquid dynamic viscosity into account. By comparing with the experimental data and other existing correlations, the proposed correlation is confirmed as an effective and concise method, and shows good accuracy in erosion prediction.