Eduardo Pereyra

Pipe-Diameter Effect on Liquid Loading in Vertical Gas Wells

The effect of pipe diameter on liquid-loading initiation has been investigated experimentally with pipes having internal diameters of 5.1(2-) and 10.2-cm (4-in.). Two-phase-flow parameters, such as pressure gradient and liquid holdup, were measured. Flow characteristics were determined by visual observation with a high-speed video camera. Critical gas-flow rate for liquid-loading initiation was identified, and comparisons between the two pipe diameters were presented. The critical superficial-gas velocity corresponding to the minimum pressure gradient was found to be faster for the smaller diameter. When the comparison was carried out in terms of mass-flow rates, critical flow rate for liquid loading in a 5.1-cm (2-in.) pipe was less than that in a 10.2-cm (4-in.) pipe. This verifies the use of velocity strings to extend the production life of the gas wells. Additionally, comparison of the data with available mechanistic-models prediction showed significant discrepancies. Possible reasons for these discrepancies are discussed.

Experimental Investigation of Three-Phase Low-Liquid-Loading Flow

Although many different studies have been conducted on gas/ liquid multiphase flow, only a very small number of three-phase flow studies, especially for low-liquid-loading flows, can be found. These studies are mainly experimental, and focused on two-phase flow in small-diameter pipelines. The coexistence of thin films of water along with oil in production systems is very commonly observed in wet-gas pipelines. The existence of the second liquid phase influences all of the flow characteristics. The three-phaseflow behavior can be considered as a combination of gas/liquid and oil/aqueous phase interactions. Meng et al. (2001) conducted two-phase-flow experiments for oil/air flow in a 2-in.-ID pipe. They observed a surprising decrease in liquid holdup and pressure gradient when the vSL was increased. They attributed this decrease to the increase in droplet entrainment. They also developed a correlation for interfacial friction factor. Fan (2005) used two experimental facilities with IDs of 2 and 6 in., respectively, to conduct two-phase water/air low-liquid-loading experiments. Fan observed stratified smooth and stratified wavy flow patterns in his experiments with the 6-in.-ID facility. With the 2-in.-ID facility, in addition to stratified flow patterns, an annular flow pattern was observed. Fan used the acquired experimental data to develop new closure relationships for mechanistic modeling. These closure relationships included wetted-wall fraction, liquid-wall friction factor, and interfacial friction factor. Later, Dong (2007) modified the 6-in.-ID facility of Fan (2005) to conduct low-liquid-loading three-phase-flow experiments. Water, air, and oil with a viscosity of 13 cp were the flowing fluids. This is a relatively high oil viscosity compared with the commonly observed values in wet-gas pipelines, and the results may not be representative for wet-gas pipeline systems. The distribution of oil and water in liquid phase for different flowing conditions was observed and categorized. In addition, a model comparison was provided for flow characteristics. Recently, Gawas (2013) used the same 6-in.-ID facility of Dong (2007) to investigate the characteristics of three-phase low-liquidloading flow. Gawas conducted his experiments by use of an oil with a viscosity of 1.3 cp for different values of water cut, and developed correlations for entrainment of liquid droplets in gas phase for twoand three-phase flows. He also analyzed the droplet-size distribution and developed a correlation for interfacial wave celerity. In addition, several studies have been conducted in other research centers to analyze low-liquid-loading flow. A summary of these studies is presented in Gawas (2013). In the current study, the facility of Gawas (2013) is used. The main objective of this research is to study low-liquid-loading threephase flow, and the targeted flow parameters are liquid holdup, water holdup, wave pattern, and pressure gradient. The experimental results for different flow characteristics are analyzed and evaluated to improve understanding of the flow phenomena. In addition, the commonly used models are evaluated by use of the acquired experimental data.

State of the Art of Experimental Studies and Predictive Methods for Slug Liquid Holdup

A summary of all available correlations and mechanistic models for the prediction of slug liquid holdup is presented. Additionally, an experimental data base for slug liquid holdup has been collected from available literature. A comparison between the predictions of available models and correlations against the data base is presented, identifying the range of applicability of the different methods. The correlations have been tuned against the new data by calculating new values of their constant parameters, showing an improved performance. Also, the uncertainties of the correlations parameters are evaluated and presented. A recommendation for the best method of predicting the slug liquid holdup is provided.

A Simplified Mechanistic Model for an Oil/Water Horizontal Pipe Separator

Summary A new methodology for oil/water horizontal pipe separator (HPS) design and performance prediction is developed. The separator di- ameter is determined on the basis of oil/water flow-pattern predic- tion. A batch separator model is adopted and modified for pipe flow to predict the separator length for achieving a desired separation quality. An experimental program is carried out to validate the pro- posed model. and the coalescence models. The hydrodynamic model considers the momentum transfer between three layers, including the water- continuous, dense-packed, and oil-continuous layers. The coales- cence submodel uses simplified population-balance equations to account for the droplet evolution within the dense-packed zone. This model requires as an input several coalescence parameters of the oil/water mixture, which cannot be obtained from a simple test. A simplified mechanistic model wherein the coalescence pa- rameters can be determined by a simple test is presented in this study. The proposed model can be used for proper design and per- formance evaluation of the HPS. Modeling Determination of two parameters are required for the HPS de- sign, namely pipe diameter (ID) and length (L). The first param- eter is defined as the required diameter to promote stratification of the phases, while the second parameter is the necessary length to achieve a desired separation quality. Because of the low liquid velocities, the momentum of the con- tinuous phase is low and gravity dominates the separation process. Thus, the behavior of the HPS can be approximated by an exten- sion of the batch-separation process. Batch-separator models pre- dict the evolution of the separation profiles as a function of time. These models are extended to the HPS by considering that both phases move at same velocity, namely the mixture velocity uM. This assumption neglects any velocity profile and momentum exchange between layers that affects the droplet coalescence. Separation Profiles. Among the available theories for batch- separator behavior prediction, the asymmetric dimple model pre-

Transient Mechanistic Model for Slug Damper/Gas-Liquid Cylindrical Cyclone (GLCC©) Compact Separator System

A dynamic model and a simulator have been developed for the Gas-Liquid Cylindrical Cyclone/Slug Damper (GLCC© -SD) system, for the prediction of its flow behavior under transient slugging flow conditions. Separate dynamic models and simulators are developed for the GLCC© and the SD units, which are integrated together with a slug generator model/simulator into an overall model/simulator for the GLCC© -SD system. Two numerical schemes are utilized for the developed integrated simulator, namely, fixed time step and variable time step schemes. Simulation examples are presented for the GLCC© , SD and integrated GLCC© -SD system, for the prediction of their performance under transient flow conditions. The GLCC© -SD simulation results demonstrate clearly the advantage of this system in dampening and smoothing the liquid flow rate under slug flow conditions, providing fairly constant flow rate at the GLCC© outlet liquid leg. The developed GLCC© -SD simulator can be extended to other separators, such as the gravity vessel separators and liquid hydrocyclones.Copyright

CFD Simulations of Low Liquid Loading Multiphase Flow in Horizontal Pipelines

Low Liquid Loading is a very common occurrence in wet gas pipelines where very small amounts of liquid flow along with the gas, mainly due to condensation of hydrocarbon gases and water vapor. The effects of low liquid loading on different flow characteristics, and flow assurance issues such as pipe corrosion prove the necessity of analyzing the flow behavior in more depth. In this study, CFD simulations are conducted for a horizontal pipe where liquid and gas are supplied at separate constant rates at the inlet. The liquid is introduced at the bottom to help shorten the developing section. The simulations are conducted with Ansys Fluent v14.5 using Volume Of Fluid (VOF) as the multiphase model. The analysis targets, mainly, the shape of the interface, velocity fields in both liquid and gas phases, liquid holdup, and shear stress profile. On the other hand, experiments are conducted in a 6-inch ID low liquid loading facility with similar testing condition. Experiments are conducted with water or oil as the liquid phase for a liquid volume fraction range of 0.0005–0.0020 of the inlet stream. For all cases, several flow parameters are measured including liquid holdup and interface wave characteristics. A comparison is conducted between CFD simulation results, model predictions, and experimental results, and a discussion of the sources of discrepancy is presented. Overall, the results help understand the low liquid loading flow phenomenon.Copyright

CFD modeling and validation for two-phase medium viscosity oil-air flow in horizontal pipes

Abstract This study presents the results of a Computational Fluid Dynamics (CFD) simulation of two-phase medium viscosity oil-air flow in a 50.8 mm internal diameter horizontal pipe. Void fraction and pressure gradient predictions were validated using experimental data for four different oil viscosities (0.039, 0.06, 0.108 and 0.166 Pa s) and different flow rates varying from 0.1 to 2.9 m/s for the gas phase and from 0.01 to 2.95 m/s for the liquid phase, where four flow patterns were predicted (stratified, dispersed bubble, bubble elongated and slug flow). The obtained results of void fraction and pressure gradient presented a mean relative error of 30.04 and 21.38%, respectively. Furthermore, the CFD results were compared against 66 empirical correlations and predictions from OLGA. It was found that between the three studied methods (CFD, OLGA and empirical correlations) the CFD model outperformed the other two methods regarding the predicted flow patterns, pressure gradients and void fractions on most cases.

Feasibility and Evaluation of Surfactants and Gas Lift in Combination as a Severe-Slugging-Suppression Method

nation angle for relatively low gasand liquid-flow rates. Sarica et al. (2014) divided the severe-slugging cycle into four steps, as described in Fig. 1. The classic pipe geometry for severe slugging is a slightly downward section upstream of a riser. In Step 1, gas and liquid velocities are low enough to allow stratified flow in the downward-sloping pipe section followed by liquid bridging and accumulation at the bottom of the riser. The hydrostatic pressure of the accumulated liquid initially increases equal to or faster than the buildup of gas pressure upstream of the liquid slug (Step 2). When the gas pressure eventually exceeds the hydrostatic head of the liquid slug, the gas will begin to push the liquid slug out of the riser and start to penetrate the riser (Step 3). The pressure in the gas reduces as the liquid is removed from the riser and the gas expands, increasing the velocities in the riser. After most of the liquid and gas exit the riser, the velocity of the gas is no longer high enough to sweep the liquid upward. Liquid film not swept from the riser starts falling back down the riser (Step 4), and the accumulation of liquid starts again. Severe slugging will cause periods of no liquid and gas production in the separator followed by very high liquidand gas-flow rates. The resulting large pressure and flow-rate fluctuations are highly undesirable. Several mitigation techniques are proposed in the literature. A thorough summary of these techniques can be found in Sarica and Tengesdal (2000). Surfactant application and gas lift are typically considered to be separate methods. The combination of both can provide a better mitigation of severe slugging by complementing one another. As mentioned by Sarica and Tengesdal (2000), Yocum (1973) was the first to identify multiple severe-slugging-mitigation techniques. These are reduction of the line diameter, splitting the flow into dual or multiple streams, gas injection into the riser, the use of mixing devices at the riser base, choking, and backpressure increase. Here, we will classify severe-slugging-mitigation methods into three groups: passive, active, and hybrids (combination of both passiveand active-mitigation methods). Passive methods require energy from the system; the most relevant are given as follows: 1. Choking: One of the most common mitigation techniques is the installation of a choke valve at the top of the riser. By choking the flow, the riser operational pressure changes, stabilizing the flow. Several publications regarding choking exist in the literature, as detailed in Sarica and Tengesdal (2000). Unfortunately, because of the backpressure created by choking, production is affected, and a minimum amount of energy is required for this method to be successful. This technique can be combined with a feedback control to regulate the largest choke opening that will stabilize the flow. 2. Backpressure increase: This method requires significant pressure increases at the separator or riser head. It is not considered to be as viable an option, even for shallow-water systems, because production-capacity reduction is experienced as a result of the backpressures imposed. The reduction in Copyright

Onset of Separated Water-Layer in Three-Phase Stratified Flow

The onset to separated water-layer in three-phase gas-oil-water stratified flow in horizontal pipes is studied theoretically and experimentally. The objective is to determine the transition boundary between the separated liquid-phase and the dispersed liquid-phase in three-phase stratified flow. An experimental facility was constructed, enabling data acquisition under horizontal three-phase stratified flow. A total of 75 experimental runs were conducted varying the water cut between 5% and 40%. The results are presented in the form of flow pattern maps, including the transition boundary between the two liquid-phase flow configurations. The transition mechanism is the occurrence of large waves at the oil-water interface, which reach the bottom of the pipe, swiping the water layer and dispersing it. A mechanistic model was developed for the prediction of the transition boundary between the separated and dispersed liquid-phase under three-phase stratified flow. The proposed model requires as input the three-phase stratified flow variables, which are determined based on the Taitel et al. (1994) model. The transition boundary is then predicted based on the proposed flow mechanism, utilizing a simple Froude number criterion. The model predictions of the transition boundary between the separated and dispersed liquid-phase flow show a good agreement with the acquired experimental data.

Design and Performance of Gas–Liquid Cylindrical Cyclone/Slug Damper System

A dynamic model and a simulator have been developed for the Gas–Liquid Cylindrical Cyclone/Slug Damper (GLCC-SD) system, for the prediction of its flow behavior under transient slugging flow conditions. Separate dynamic models and simulators are developed for the GLCC and the SD units, which are integrated together with a slug generator model/simulator into an overall model/simulator for the GLCC-SD system. Simulation examples are presented for the GLCC, SD and integrated GLCC-SD system. The GLCC-SD simulation results demonstrate clearly the advantage of this system in dampening and smoothing the liquid flow rate under slug flow conditions, providing a fairly constant flow rate at the GLCC outlet liquid leg.