PSI - Issue 71

Rahul Tarodiya et al. / Procedia Structural Integrity 71 (2025) 241–247

242

1. Introduction Surface erosion due to repetitive impact of particles is one of the major causes of reduction in the performance and service life of the components handling solid particulate flows. In the process of oil and gas extraction from wells, small sand particles in the fluid impacting the pipeline system and thereby causes erosive wear failure of the pipeline and fittings, such as elbows, plug-tees, pumps, and valves (Tarodiya and Levy 2021; Tarodiya and Gandhi 2017). Complete elimination of the surface erosion of the components is difficult; however, it can be reduced by identifying the causes, which may help to optimize the operations of the process and timely maintenance of the system. In recent years Computational Fluid Dynamics (CFD) based methodologies are being commonly adopted to quantify the erosion in pipelines and fittings. The estimation of surface erosion by using CFD based approaches is performed through tracking the particle motion, collecting the particle-wall collision data, and computing the material loss rate by using suitable erosion correlation (Tarodiya et al. 2022). Tarodiya and Levy (2023) conducted CFD simulations to investigate the location of maximum erosion in a single 90-degree bend for gas-solid flows and proposed a correlation to estimate the location of maximum erosion. Similarly, many works have been conducted to numerically investigate the erosion behavior of the single elbow (Edward et al. 2001; Solnordal et al. 2015; Peng and Cao, 2016; Laín and Sommerfeld, 2019). In a real pipeline network, elbows are often mounted in series and the distance between the two elbows may be very short. The upstream flow could affect the erosion of the elbow in series and thus the erosion behavior of the elbow in series is more complex as compared to that of a single elbow. Investigators (Kumar et al., 2014; Asgharpour et al. 2017; Othayq et al. 2021) conducted experimental studies to investigate the erosion of elbows in series. They reported higher erosion of the primary elbow as compared to secondary elbow in series. However experimental studies are limited, due to the requirement of a long time to get the measurable amount of material loss, high cost involved and difficulties in performing experiments. In view of complexity, CFD based approaches are being used to develop an understating on the erosion of the elbows in series. Felten (2014) numerically investigated the effect of connecting length on the erosion of the secondary elbow. Sand particles of uniform size of 180 µm are being used to conduct the analysis. They argued that for the connecting length greater than 24D (L/D>24), the maximum erosion rate is independent of the connecting length. Zhao et al. (2022) used π shaped geometry to investigate the erosion of elbows mounted in series with sand particles of size 50 µm and 200 µm. They studied the effect of bend orientation, curvature radius, and connecting length on the erosion of the elbows in series. They reported that the erosion of the downstream elbow is greatly influenced by the upstream connecting pipe length. It has been found through the literature that the erosive wear behavior of elbows mounted in series is complex. Design parameters, such as connecting length between the two elbows is one of the critical parameters affecting the particulate flow trajectory which could influence the erosive wear behavior of the secondary elbow. Studies are conducted, both experiment and numerical, to determine the effect of connecting length on the erosion of the secondary elbow. However, the understanding developed is limited to the L/D ratio up to 32 and with only one particle size range. Since the particle size could affect the particle trajectory, thus, the understanding of erosion in elbows with the variation in connecting length between two elbows needs to be investigated for the flow with different mean size particles. The present paper aims to numerically develop an understanding of the variation in connecting length between two elbows on the erosive wear behavior of the elbows at different particle sizes and flow velocities. 2. Numerical Modeling The prediction of erosion rate using Computational Fluid Dynamics (CFD) involves several key steps: simulating the flow of the continuous phase, tracking particles, and applying an erosion model. This study utilizes the Eulerian Lagrangian approach to simulate the gas-solid flow. The gas phase is treated as a continuous medium, with the Reynolds-averaged Navier-Stokes (RANS) equations employed to obtain solutions within the Eulerian framework. Solid particles are considered as a discrete phase, and their motion is tracked in a Lagrangian framework using Newton’s second law. An erosion model then calculates erosion at specific locations based on the particle -wall collision data. A two-way coupling approach is used to account for the interactions between the solid and gas phases. Solid particles reflect back into the fluid medium after colliding with pipe walls, and their subsequent paths are determined using a particle rebound model. This study applies the Grant and Tabakoff (1975) particle rebound model, which has been validated in the literature for its effectiveness in CFD modeling for erosion prediction. The normal (e n ) and tangential (e t ) restitution coefficients in this model depend on the particle impact angle (α) and is given by. 2 3 n e 0.993 1.76 1.56 0.49 = − +  −  (1) 2 3 t e 0.988 1.66 2.11 0.67 = − +  −  (2) 2.1. Geometry and meshing