A numerical strategy to account for the effect of self-induced geometry changes in wear estimation

The erosion of a surface impinged by solid particles dragged by a liquid is a serious concern in the oil&gas industry. Recent experiments indicate that the changes in the geometry produced by the erosion may significantly alter the evolution of the erosion process itself. This typically happens when a surface is exposed to aggressive flow condition for a prolonged period, and it becomes a significant issue for complex devices such as valves. However, none of the predictive techniques available at present, which mainly involve the use of algebraic erosion correlations in conjunction with an Eulerian-Lagrangian model, appears capable in handling this phenomenon. In this study, we propose a numerical strategy that allows accounting for the effect of the self-induced geometry changes on the wear estimates. The evolution of the erosion process is estimated by an effective post-processing technique applied to the solution of a steady-state Eulerian-Lagrangian simulation. The good agreement between our predictions and experimental data available in the literature regarding an abrasive jet impingement test makes the application of this strategy to more complex flows very promising.