A new correlation for estimating solid particle erosion in elbows

Abstract

Throughout the Literature, solid particle erosion prediction models have been developed based on limited data without considering uncertainties involved in measurements. In this study, a new correlation is designed to estimate the maximum erosion rate caused by solid particles in standard elbow geometries. The novel development process of this correlation includes uncertainty estimation of erosion measurements. To estimate uncertainties in an erosion database, it is imperative to have a model that can be used to calculate accurate erosion values and provide reliable references. For this purpose, a database of repeated erosion measurements for gas-sand, low-pressure conditions collected by previous researchers is used. The database includes standard elbow geometries with 0.0762 m and 0.1016 m internal pipe diameter, and sand particles with mean diameters of 75 and 300 μm. Then, the erosion values in this database are adjusted to agree with the physical behaviors expected in the erosion process. Finally, based on the adjusted erosion values, a new correlation is developed that accounts for important parameters such as pipe and particle diameters, and gas velocity. The results and comparison of this new model with other correlations/models in the literature show that this correlation performs very well in predicting solid-particle erosion values for elbow geometry with an RMSE value of 7.96E-02 mm/kg. The effect of superficial liquid velocity is also investigated and modeled based on a database of gas-liquid-sand erosion, resulting in an updated version of the correlation for predicting erosion in gas-liquid-sand flow conditions. Furthermore, CFD results for higher-pressure fluids and larger pipe diameters of 0.1524 m and 0.2032 m and mean particle sizes of 25, 75, 150, and 300 μm, are used to further validate the correlation for larger pipe sizes and high-pressure fluid flows. In summary, three correlations have been developed in this research; one for low-pressure flows with application in uncertainty estimation, one for predicting erosion in two-phase flow conditions, and one for cases with larger pipe diameters or high-pressure fluids.