Numerical prediction of erosion distributions and solid particle trajectories in elbows for gas-solid flow

Erosion caused by particles in pipe bends is a serious problem in the oil and gas industry, which may cause equipment malfunction and even failure. The majority of this work studies the particle trajectories and erosion distributions in pipe bends under different influencing factors by using the computational fluid dynamics (CFD) method. A two-way coupled Eulerian-Lagrangian approach is employed to solve the gas solid flow in the pipe bend. Eight commonly used erosion models and two particle-wall rebound models are combined to predict the erosion rate on the 90 elbows. The Det Norske Veritas (DNV) erosion model with the Forder et al. particle-wall rebound model is finally chosen as a sample to develop the new CFD-based erosion model after comparing with the experimental data. The accuracy of this presented model is assessed by the experimental data available in previous literature for a range of flow conditions. Good agreement between the predictions and experimental data is observed. Further, the erosion distributions and particle trajectories in pipe bends under different flow velocity, particle mass flow rate and mean curvature radius to diameter (R/D) ratio and pipe diameter are investigated by applying the presented model. The results show that totally two types of erosion scars and three types of particle collisions occur at the elbows with different erosion parameters. These two types of scars may occur alone or occur together due to the combined effect of the particle collisions. Finally, two equations for predicting the maximum erosion location are obtained considering the pipe bend orientation, the particle diameter and the R/D ratio. (C) 2016 Elsevier B.V. All rights reserved.