Electrical submersible pump (ESP) technology has been widely applied in the oil and gas industry due to its high productivity. However, erosion always causes the reduction of productivity and sometimes even the failure of an ESP system. This study explores the effect of gas presence on erosion mechanism on an ESP which is composed of 4 stages of Helico-Axial Pump (HAP). A 200-hour erosion test has been performed on this ESP. During the test, the ESP was running at 3600 RPM with a liquid flow rate of 880 GPM, 20% inlet Gas Volume Fraction (GVF), and 0.24% sand concentration by weight. Performance tests were conducted every 50 hours to acquire the performance maps and monitor the performance degradation. Analysis of volume/weight loss and performance degradation is conducted to investigate pump wear. Two types of erosion are found at the impeller: the volume loss found notably at the leading edge is mainly caused by two-body impact erosion, while the tip clearance increment between the impeller housing and impeller blade tip is mainly caused by the three-body abrasive erosion. Unlike most conventional centrifugal pumps, there is no observable wear found at the trailing edge of the impeller. The presence of the gas shows a negative effect on both types of erosion. The consequence of the erosion is the performance degradation, especially at the condition with higher pressure rise. It is suggested to apply this HAP in the oil field with more gas and higher bottom hole pressure.

The concentration of the sand inside the flowlines can directly affect the magnitude of erosion in various pipelines and pipe fittings components. The erosion models available in the literature suggest that the relationship between erosion and sand concentration is linear. Furthermore, some authors believe that for a volume concentration of sand approximately below 0.75% by volume (corresponding to about 2% by mass in liquid), the effects of sand concentration on the rate of erosion are minimal. However, one could ask how this limitation on the amount of sand concentration varies in various geometries as the local concentration at the location where the sand particles are impacting can change within a flow geometry and how this can affect the maximum local erosion rates. Therefore, the sand concentration distributions inside a vertical pipe and two elbows in series for liquid-solid and liquid-gas-solid flows are examined in this investigation by CFD simulations and experimentally. Experiments are performed in a facility with two vertical pipe sections (101.6 mm reducing to a 50.8 mm inner pipe diameters) and immediately after the first and second elbows (50.8 mm inner pipe diameter) from the inner to the outer diameter of the elbows. The concentration distributions are associated with erosion measurement results in both the first and the second elbows. Finally, the CFD simulations of the corresponding experiments in liquid-sand and liquid-gas-sand flows are conducted and compared with the experimental erosion patterns and magnitude as well as various concentration distributions.

Large diameter pipes and elbows are vastly used in industry especially in mining and oil and gas production. Solid particle erosion is a common issue in these pipelines, and it is important to predict it to avoid failures. Currently, laboratory experiments reported in the literature are limited to diameters less than 4 inches. Therefore, there is not much experimental data available for large diameter elbows. However, the erosion can be predicted by CFD simulations and applying erosion equations in such elbows. The goal of this project is to examine the effects of elbow diameter and Stokes number on erosion patterns and magnitude for various flow conditions for elbow diameters of 2, 4, 8, and 12 inches. The approach of this work is to first perform CFD simulations of liquid-solid and gas-solid flows in 2-inch and 4-inch elbows, respectively, and evaluate the results by available experimental data. Then CFD simulations are carried for 2, 4, 8, and 12-inch standard elbows for various Stokes numbers corresponding to gas dominant flows with the velocity of 30 m/s, and liquid dominant flows with the velocities of 6 m/s. For gas dominant flows erosion in air and for liquid dominant flows erosion in water is investigated. All these simulations are carried for four particle sizes of 25, 75, 150, and 300 microns. The results indicate that Stokes number and diameter of elbows have significant effects on erosion patterns as well as magnitudes in this geometry. This work will have various applications, including validating mechanistic models of erosion predictions in elbows and developing an Artificial Intelligence (machine learning) algorithm to predict erosion for various flow conditions. Such algorithms are limited to the range of conditions they are trained for. Therefore, it is important to expand the database these codes are accessing. Overall, the CFD database of large diameter elbows will reduce the computational costs in the future.

Solid particle erosion is one of the most commonly encountered problems faced by the oil and gas industries during production and transportation processes. The severity of solid particle erosion is affected by multitude of factors such as particle properties, fluid flow properties and flow geometry, flow regime, and target material properties. The present work investigates the effect of particle size on solid particle erosion in gas flows. Sharp quartz particles with nominal sizes of 75, 150, 300 and 600 μm are used in this work. Particle Image Velocimeter (PIV) is used to measure the particles velocities distributions for various particle flow rates. An average particle velocity of 24 m/s is used to conduct erosion experiments for various particle sizes and two particle rates on Stainless Steel 316 at two different impact angles of 15 and 90 degrees. Comparison of measurements for two particle flow rates of approximately 0.02% and 0.002% by volumes demonstrates that increased particle flow rate can affect the carrier fluid’s flow field and change particle velocities within the carrying fluid. In the erosion experiments, the magnitude of erosion ratio increases as there is an increase in particle size. A preliminary erosion model is developed that can be used in CFD simulations of solid particle erosion for various particle sizes.

The performance of multi-stage Electrical Submersible Pumps (ESPs) under different flow conditions and its life span with sand production are commonly predicted by the Computational Fluid Dynamics (CFD) simulations. The mesh generation methodology and optimum grid number are usually validated by pump water catalog curves. Then, the validated mesh geometry is adopted in high viscosity, multiphase flow, and sand erosion simulations to study the effects including but not limited to: discrete phase bubble diameter, turbulence model, body forces, and erosion models. However, the mesh validation by pump water curves is not enough in complex flow conditions, especially in the erosion simulations. Different from the pump hydraulic performance simulation, the accuracy of the erosion simulation can be affected by mesh boundary and inner layer grid thickness, especially for small particles. In addition, the mesh-type (hexahedral and tetrahedral) and size of the inner domain can also significantly affect the particle trajectory. A comprehensive mesh independent study is conducted for water, oil, and gas-liquid conditions of a mixed type ESP in this paper. Then the near-wall inflation layer thickness and inner domain grid size effect to ESP erosion simulation are well analyzed. The mesh generation methodology can be applied to other turbomachinery simulations to improve accuracy.

AI approaches include machine learning algorithms in which models are trained from existing data to predict the behavior of the system for previously unseen cases. Recent studies at the Erosion/Corrosion Research Center (E/CRC) have shown that these methods can be quite effective in predicting erosion. However, these methods are not widely used in the engineering industries due to the lack of work and information in this area. Moreover, in most of the available literature, the reported models and results have not been rigorously tested. This fact suggests that these models cannot be fully trusted for the applications for which they are trained. Therefore, in this study three machine learning models, including Elastic Net, Random Forest and Support Vector Machine (SVM), are utilized to increase the confidence in these tools. First, these models are trained with a training data set. Next, the model hyper-parameters are optimized by using nested cross validation. Finally, the results are verified with a test data set. This process is repeated several times to assure the accuracy of the results. In order to be able to predict the erosion under different conditions with these three models, six main variables are considered in the training data set. These variables include material hardness, pipe diameter, particle size, liquid viscosity, liquid superficial velocity, and gas superficial velocity. All three studied models show good prediction performances. The Random Forest and SVM approaches, however, show slightly better results compared to Elastic Net. The performance of these models is compared to both CFD erosion simulation results and also to Sand Production Pipe Saver (SPPS) results, a mechanistic erosion prediction software developed at the E/CRC. The comparison shows SVM prediction has a better match with both CFD and SPPS. The application of AI model to determine the uncertainty of calculated erosion is also discussed.

The effects of particle size on erosion magnitude and erosion profiles are investigated experimentally in a submerged slurry jet impingement facility. The slurries were diluted to avoid concentration effects on the flow field and the resulting erosion. The experiments are performed with particle sizes of 25, 75, 150, 300, and 600 μm. Experimental results demonstrate different erosion severity and pattern for the various particle sizes. It is critical to have a reliable and accurate tool to predict erosion for different particle sizes. Previously, a comprehensive CFD-based procedure to predict erosion for various particle sizes was proposed by investigators at the Erosion/Corrosion Research Center (E/CRC). A feature of this procedure is that it can account for particle size in more detail than previous methods. In this study, the computational procedure is applied to conditions of the present experimental data. Particle impact parameters are extracted to explain the effect of particle size on the resulting erosion. The predicted results are compared with data which demonstrate possible shortcomings of the available CFD based techniques for predicting solid particle erosion. The results indicate that with proper use of grid spacing near the wall, the CFD-based erosion calculation method with existing erosion models can predict the trend of the experimental data, though improvements are still needed to the models to accurately account for particle size effects.

During production of oil and gas from wells, solid particles such as removed scales or sand may accompany petroleum fluids. These particles present in this multiphase flow can impact inner walls of transportation infrastructure (straight pipelines, elbows, T-junctions, flow meters, and reducers) multiple times. These repeated impacts degrades the inner walls of piping and as a result, reduce wall thickness occur. This is known as solid particle erosion, which is a complex phenomenon involving multiple contributing factors. Prediction of erosion rates and location of maximum erosion are crucial from both operations and safety perspective. Various mechanistic and empirical solid particle erosion models are available in literature for this purpose. The majority of these models require particle impact speed and impact angle to model erosion. Furthermore, due to complex geometric shapes of process equipment, these solid particles can impact and rebound from walls in a random manner with varying speeds and angles. Hence, this rebound characteristic is an important factor in solid particle erosion modeling which cannot be done in a deterministic sense. This challenge has not been addressed in literature satisfactorily. This study uses experimental data to model particle rebound characteristics stochastically. Experimental setup consists of a nozzle and specimen, which are aligned at different angles so particles impact the specimen at various angles. Information regarding particle impact velocities before and after the impacts are obtained through Particle Tracking Velocimetry (PTV) technique. Distributions of normal and tangential components of particle velocities were determined experimentally. Furthermore, spread or dispersion in these velocity components due to randomness is quantified. Finally, based on these experimental observations, a stochastic rebound model based on normal and tangential coefficients of restitutions is developed and Computational Fluid Dynamics (CFD) studies were conducted to validate this model. The model predictions are compared with experimental data for elbows in series. It is found that the rebound model has a great influence on erosion prediction of both first and second elbows especially where subsequent particle impacts are expected.

Empirical wear coefficients are used in concert with computational fluid dynamics (CFD) codes which model highly loaded slurry flows through centrifugal pumps in order to predict erosive wear in wet-end components. In practice, separate sets of wear coefficients are used to determine the contributions of sliding and impact wear to the total erosive wear at the wetted surface. In this study, experiments were performed in order to obtain the impact wear coefficients for sand in a water slurry impinging on high chrome white iron alloys that are commonly used in the construction of heavy duty centrifugal slurry pumps. Two separate sets of tests were completed using two different types of narrowly graded sand, with mass median particle diameters of approximately 600 μm and 300 μm respectively. The tests were performed in a closed loop containing a slurry pump, flow meter, inverted U loop for determination of the solids concentration, and 8 sample sections arranged serially. Each sample section was constructed from polyurethane, with rectangular flow cross sections of 1 inch (25.4 mm) width and 2 inch (50.8 mm) height and lengths of 1 foot (305 mm). One metal sample was placed into each sample holder so that it spanned across the 1 inch (25.4 mm) width and was exposed to the slurry flow, with its edges being supported by the flat polyurethane walls on either side. The samples were machined to have constant angles on the leading edge faces which varied from 10 to 60 degrees (from sample to sample), in order to obtain a range of impact angles (angle between the particle trajectory and the wetted surface) of the particles impinging on the sample leading edge faces. Tests were run at 12 % concentration by volume and at mean channel-sectional flow velocities of 10 m/s, with run times varying from 30 minutes to 180 minutes over the course of the test program. Slurry loop samples were taken at the beginning and end of each run in order to determine the particle size distribution and to monitor degradation of solids through sieve and micrograph analysis. The worn wedge face surfaces were scanned at intermittent times throughout the testing using an optical profilometer, and the local erosive wear was determined on the slanted face of, as well as at the tip of, the wedge-shaped samples. The progression of wear over the course of the test program was measured and analyzed in this manner. The local solids concentration, velocity, and impact angle was then predicted using in-house CFD codes formulated in the same manner as the pump wear models. The experimental wear profiles, together with the predicted local solids concentration, velocity, and impact angle, were then used to calculate the specific energy coefficient (or impact wear coefficient) at multiple impact angles. A formulation for the impact wear coefficient as a function of impact angle at a given particle size was then produced at each of the two different particle diameters. By comparing the data between the two different particle diameters, an adjustment factor for particle diameter was then formulated. This paper primarily focuses on the experimental test program, providing a description of the experiments, results, and data analysis, as well as a discussion of the results and some description of the test-derived wear coefficient formulations.

The term 1/2ρV 2 represents kinetic energy per unit volume; it is also known as dynamic pressure, or velocity head, in flowing fluids and has units of pressure. It is one of the primary parameters in many wide-ranging phenomenon of practical interest in control valves and valve-related systems. The phenomena of interest include vibration, noise, solid particle erosion, liquid impingement erosion, cavitation and more. The description of the extent of these phenomena through their physics, as relating to acceptability of the system performance, is closely tied to the magnitude of the kinetic energy of the flowing stream; this permits use of this parameter as a criterion for reliable performance of control valves and valve-related systems. Kinetic energy criterion is a more general approach, when compared to the older engineering practice of specifying velocity limits only, because it takes into account the effect of fluid density which can be different by orders of magnitude depending on the process. The robustness of this criterion is confirmed by the results of its application in solving valve problems. Its main advantages are simplicity while staying close to the phenomena of interest.

Solid particle erosion is a complex phenomenon that depends on many factors such as particle and flow characteristics, type of material being eroded and flow geometry. Bends, which are integral parts of piping system, are vulnerable in erosive environments because solid particles can not follow the streamlines of the flow and impinge on the pipe wall. Thus, this paper investigates numerically the erosion phenomenon that occurs in 90° and 180° curved ducts. The numerical calculations are performed by Eulerian approach for gas-phase taking into account the mutual effects of the solids on the gas and Lagrangian approach for dispersed-phase. The effects of particle rotation and lift forces are included in the particle tracking model while the effect of inter-particle collisions is neglected. The erosion prediction model comprises three stages: flow modeling, particle tracking and erosion calculations. Many of the previous published data are taken to validate the three stages of the present model. Comparisons between predicted penetration rate and published experimental data show a good agreement. The effects of bend orientation, inlet gas velocity, bend dimensions, loading ratio and particle size on the penetration rate are also simulated. The present results show that the penetration rate increases as the curvature ratio, particle diameter, inlet gas velocity increase and as the mass loading ratio and pipe diameter decrease. Furthermore, the present results show also that the bend orientation and flow direction have no significant effect on the penetration rate.