This document is an expanded abstract.
Summary
Solid particle erosion is considered as one of the major concerns in engineering applications which can lead eventually to failure of components and potentially result in hazardous catastrophic consequences. This work presents a CFD approach for erosion prediction applicable for gas-solid and liquid-solid flows. Reynolds-averaged Navier-Stokes (RANS) equations with standard k–e model are used to predict fluid flow. Discrete phase model (DPM) combined with Forder et al. particle-wall rebound model is employed to predict particle motion. Comparing to two-way coupling ways, one-way between flow and particle motion is sufficient for the present purpose. A few erosion models, including that of Oka, were tested. In order to validate the current approach, predictions are compared to experimental results in the existing literature. The Oka erosion model gives prediction closest to experimental data.
Introduction
In many engineering internal flows, solid particles are often carried by flowing fluid. For example, particles flowing through oil and gas pipelines, turbines, choke valves. The particles impinge on the surface of these components and chip off material from the surface. The surface is said to be eroded. More technically, removal of surface material by continuous impingements of abrasive particles transported in flowing fluid is called solid particle erosion [1]. Those eroded components are common in industrial operation. They face a high risk of failure and potentially cause hazardous catastrophic consequences.
Research has revealed the complexity of particle erosion process. As many as 33 parameters including flow conditions, impingement velocity, angle, particle size and distribution are identified to affect erosion [2]. In order to mitigate erosion problems, a good understanding of erosion is required. Such understanding can be obtained through lab-scale experimental studies or numerical modeling. Experimental studies, although provide excellent insights into physics of erosion process, are costly and time-consuming. Comparatively, numerical prediction, which is developed on the basis of experimental understandings and rapid development of computer science, provides a good complementary understanding of erosion process. This study presents a Computational Fluid Dynamics (CFD) model of erosion modeling in gas-solid and liquid-solid flows.
