ABSTRACT Design and operation of gas-condensate pipelines presents many challenges to the flow assurance engineer and pipeline operator. Important challenges are accumulation during low production and the removal of liquid during a ramp-up operation. Previously the connection between the behavior during a ramp-up scenario and multiple solutions for the steady-state fully-developed approximation was shown in [1],[3]. In this paper we will study the behavior during a ramp-down scenario where a pipeline or a well starts to accumulate liquid. The focus will be on understanding the phenomena. We will show how the steady-state fully-developed approximation can predict the behavior of a full transient simulation and we will make comparisons with laboratory experiments for different ramp-down scenarios. 1 INTRODUCTION Simulation of multiphase gas-condensate pipeline transport with low liquid loading is a challenging task. Important properties are the pressure drop for high rates and liquid accumulation at low rates which contribute to determining the operational envelope of the field.

ABSTRACT After the first oil crisis in 1973 the gas/oil industry shifted its focus on producing oil and gas outside the Middle East. This triggered offshore activities with the need to transport oil and gas simultaneously in pipelines over great distances to shore. At that moment a lot had been achieved on multiphase flow in production tubing, triggered by the pioneering work in 1963 by Duns and Ros with their flow-pattern dependent model. However, the multiphase transport in near-horizontal pipelines was very much more complex, since gravity was now working perpendicular to the flow. The Beggs-Brill (1973) paper with multiphase flow measurements and a widely used method for horizontal and inclined pipes offered interesting starting points. Also their course on a broad range of correlations and their validation with extensive sets of field data was most welcome for new comers in the field like me. The development of the methods with considerable improvements in the physics, is very nicely summarized by Shippen and Bailey (2012) in the figure shown above. In my opinion the modelling period started already in 1963 with the publication of the first flow-pattern dependent model accounting for slip between the phases. Field data collected during his stay at Shell Oil in the USA enabled Nico Ros to adapt his closure relations based on atmospheric laboratory data at Shell Rijswijk. This led to an important improvement of the accuracy of his model.

ABSTRACT Understanding the dynamics of multiphase flows to enable their accurate and efficient prediction is of central importance in the oil-and-gas industry. Despite the significant advances that have been made in modelling these flows over a number of decades, however, numerous challenges, and open problems, remain. We present examples of progress made recently in the Multi-scale Examination of MultiPHase physIcs in flowS (MEMPHIS) programme aimed at providing reliable predictive tools. We also highlight the limitations of purely CFD-based approaches, and provide our perspective on potential predictive strategies based on a ‘multi-fidelity’ approach: a true fusion between model-based and data-driven modelling. 1. INTRODUCTION Multiphase flows and associated phenomena are ubiquitous in the oil-and-gas industry. As highlighted by the work of Shippen and Bailey [1] "a monumental amount of research and development work has been invested in multiphase flow modelling over the past 50 years." According to the classification of Brill and Arirachakaran [2], we have witnessed a transition from a purely "empirical period" from 1950–1975, which relied on the exclusive use of correlations, to the present "modelling years" era that started in the 1980s wherein multi-fluid models gained prominence; the latter combined momentum conservation equations and flow regime transitions. These models incorporated force balances, accounted for the need to satisfy certain stability and well-posedness criteria, but, nonetheless featured numerous empirical, closure relations [3].

Abstract Heavy oil production and transport remains a challenge due to the lack of understanding of the flow behavior of gas and high viscosity oil two-phase flow in pipes. For example, existing two-phase flow in pipes mechanistic models shows significant discrepancy in predicting high-liquid-viscosity two-phase flow pattern transitions in horizontal and slightly inclined pipes. The objective of this study is to improve the modeling of stratified-smooth/ stratified-wavy flow pattern transition in horizontal and slightly inclined pipes for high-liquid-viscosity two-phase flow application. This objective is achieved by proposing a liquid viscosity-dependent sheltering coefficient model into the Taitel and Dukler (1976) stratified-smooth/stratified-wavy transition model. Further objective is to propose a new slug/annular transition criterion in horizontal and slightly inclined two-phase flow applicable for high-liquid-viscosity two-phase flow. A validation study of the proposed transition models is carried out against experimental high-liquid-viscosity two-phase flow pattern data sets. The validation result revealed that the proposed stratified-smooth/ stratified-wavy, and the slug/annular transition models predict the high viscosity horizontal flow pattern data more accurate than that of the original Taitel and Dukler (1976) model.

Abstract This paper discusses a transient solid-liquid two-phase flow modelling approach that was applied for real-time drilling hydraulic and cuttings transport calculation. The flow patterns are categorized into seven scenarios, and the flow pattern transition criteria are presented. A generalized transient solid transport model is developed to simulate the transient solid-liquid two-phase flow by combining the transient mechanistic models and the flow pattern prediction method. Introduction Hole cleaning or cuttings transport is one of the major challenges in drilling, especially for extended reach drilling (ERD). The physics behind cuttings transport is solid-liquid, two-phase flow in annulus geometries. A closely related research topic is the solid-liquid, two-phase pipe flow. A critical part of the two-phase flow study for both pipe flow and annulus flow is the understanding of the flow patterns. In engineering applications, a common approach for two-phase flow modelling is to predict the flow pattern and then develop mechanistic models to simulate the flow based on the flow patterns. Accurate flow pattern is the fundamental of the mechanistic models. The predictions from the model may be completely wrong if the flow pattern is not correct.

Abstract In oil and gas production, annular flow is a common flow regime found in wells and pipelines. Predicting erosion in multiphase flow is a challenging task as so many factors and phase interactions are involved. Computational Fluid Dynamics (CFD) offers a way to predict multiphase flow erosion. This present work shows how this state-of-the-art erosion model is applied to a 3 inch elbow to calculate erosion under annular flow conditions and how an improved 2-D model is developed for calculating erosion in annular flow for elbow geometries. The CFD predicted results are compared with experimental data and good agreement is observed. Flow solution from CFD and collected erosion data are also utilized to improve a 2-D model for annular flow application. It is shown that the combined CFD and 2-D model is a promising erosion prediction procedure for annular flows. Introduction Predicting solid particle erosion in multiphase flow is difficult as so many factors are involved in the problem. Currently, both experimental and numerical approaches can be applied to investigate the phenomenon of multiphase flow erosion. Several experimental studies were conducted at the University of Tulsa Erosion/Corrosion Research Center (E/CRC) to measure erosion in multiphase flow (Dosila, 2008; Vieira, 2015; Parsi; 2015; Fan, 2010). However, there are still many questions unanswered and especially modelling of multiphase flow erosion has not been extensively studied. This is especially the case for annular flow commonly found in oil and gas production. Dosila (2008) found through experimentation that for annular flow the erosion rate can decrease when liquid flow rate increases above a critical value. 2 Fan (2010) also observed the same behavior even for large pipe diameters. 3 Vieira (2015) studied the effect of elbow orientation on erosion in annular flow. He found that erosion in a vertical-horizontal elbow was significantly higher than that in a horizontal-horizontal elbow. He improved a 1-D simplified model by increasing the initial particle tracking velocity. The factor was obtained empirically through flow experimental data 4 .

Abstract Most multiphase flow models are developed and validated using a limited set of experimental data. Even when extensive experimental data are available, they are collected at laboratory conditions and/or test facilities, whose operating ranges and scales do not usually coincide with the field conditions. Therefore, models are routinely used for extrapolations to conditions where experimental data is not available. For example, this is often the case for very remote operations such as deep water applications. In this paper, we present three approaches to quantify experimental data and model prediction uncertainty for multiphase flow applications. We demonstrate these methods using three applications, estimation of experimental data uncertainty of erosion extent measurements, quantification of prediction uncertainty of critical velocity for sand transport, and of liquid entrainment in two-phase flow. Introduction Multiphase flow occurs in many systems encountered in chemical, petroleum, and nuclear industries. Because efficient design and operation of these systems depend on the ability to predict the flow characteristics, several multiphase flow models have been developed. These models are based on mechanistic approaches, which utilize the conservation equations of mass, momentum, and energy along with empirical or semi-mechanistic closure relationships and adjustable parameters. The resulting equations are solved using various numerical methods and algorithms. The inputs to these models include fluid properties such as density, viscosity and surface tension, flow geometry such as diameter and inclination of the pipeline, and operating conditions such as superficial gas and liquid velocities. The outputs may be flow patterns, pressure gradients, liquid hold-up, etc.

Abstract Several semi-mechanistic and empirical models exist in literature for solids transport in single-phase carrier flows in pipes and channels, which could be extended to fully-mixed multiphase carrier flows (e.g., bubbly flows). Very few attempts have been reported for solids transport in segregated multiphase flows (e.g., stratified flows). Building on the previous works of Oroskar and Turian (1980) for solids transport in single phase carrier flows, this paper proposes a new semi-mechanistic model for solids transport in pipes taking into account the multiple length scales that exist in stratified multiphase (liquidgas) flow scenarios. Comparisons of the proposed model predictions to experimental data and existing models are discussed.

Abstract The gas drift velocity in an elongated bubble can be measured as the bubble velocity moving through stagnant liquid in a pipe. In this study, Computational Fluid Dynamics (CFD) is used to numerically simulate the motion of elongated gas bubbles into liquidfilled channels and pipes. The steady, inviscid flow CFD solution agrees with the analytical solution. Furthermore, the CFD solution for viscous flow agrees with new experimental data. Two flow regimes were predicted by the viscous flow simulations: one of constant bubble velocity and another with decreasing bubble velocity over time. A change in flow regime is observed both in terms of the bubble shape and the gas drift velocity. Correlations are derived from the CFD results that describe the time dependent drift velocity as a function of the liquid viscosity.

ABSTRACT: In experimental studies using polymeric type drag reducers, mechanical degradation is often a problem. A method is described, based on partially degrading the solution to get a fluid that is stable over time, which greatly simplifies working with DRA. The method has been adopted to demonstrate, and quantify, diameter effects in single-phase flow. Further, the effects of DRA in two- and three-phase flow have been studied in a D=100mm pipe. The overall conclusion is that the drag reduction increases with the inlet fraction of the liquid that carries the DRA. INTRODUCTION This paper describes some key findings from experimental work on the effects of polymeric type drag reducing agents (DRA) in multiphase pipe flow, carried out at IFE. The work has supported the development of a mechanistic model that accounts for the effects of DRA in multiphase flow, a work that is hindered by the yet unresolved diameter dependence in the friction law. DRAG REDUCERS – A BRIEF LITERATURE OVERVIEW The reduction of drag (or pressure drop) in turbulent flows by adding long-chained, high molecular weight polymers was first reported by Toms in 1948. Today, chemical type drag reducer agents are in two classes; the polymer type (PDRA) and the surfactant type drag reducers (SDRA). Only PDRA are dealt with in this paper. The PDRA suffer from permanent mechanical degradation (polymer scission) in pumps and valves, resulting in reduced performance, and are therefore mostly used in once-through applications. The SDRA, on the other hand, are capable of forming very long cylindrical micelles, with self-healing properties. SDRA are therefore more suitable for closed circulating fluid systems such as district heating pipelines and heat exchangers. The oil and gas industry have traditionally used drag reducers to reduce the pressure drop for the transport of single-phase liquids over long distances.

ABSTRACT: This paper presents the results of three uncertainty propagation and quantification methodologies, i.e., perturbation method, Taylor series expansion based approach, and Monte Carlo simulation, for deterministic multiphase fluid flow models. The sensitivity of pressure drop and liquid holdup predictions to slug length and liquid entrainment in gas core uncertainties is investigated for a multiphase fluid flow model, TUFFP Unified Model (1). The analysis shows that the Taylor series expansion consistently overestimates the output uncertainty for all cases when compared to perturbation method and the Monte Carlo simulation results. The results of perturbation method are similar to Monte Carlo simulation results. 1 INTRODUCTION Multiphase flow occurs in many systems encountered in chemical, petroleum, and nuclear industries. Because efficient design and operation of these systems depend on the ability to predict the flow characteristics, several multiphase flow models have been developed. These models are based on mechanistic approach, which utilizes the conservation equations of mass, momentum, and energy along with the empirical or semi-mechanistic closure relationships and adjustable parameters. The resulting equations are solved using various numerical methods and algorithms. The inputs to these models include fluid properties such as gas and liquid densities, viscosities, and surface tensions, pipeline geometry such as diameter and inclination angle, and operating conditions such as superficial gas and liquid velocities. The outputs are flow patterns, pressure gradient, liquid holdup, and flow characteristics. The multiphase flow models are used extensively to design and operate multiphase flow pipelines with considerable success. However, the predictions from these models can occasionally differ up to 100% compared to field data. We argue that part of this discrepancy is due to the inherent uncertainty in these models. However, these models are not equipped to represent, propagate, and quantify the resulting uncertainty in the model predictions, because they are deterministic.

ABSTRACT This paper presents a novel concept of measurement in the area of single and multi phase flow instrumentation to build in real-time the two dimensional cross section of a fluid passing through the pipe. The principle of operation consists of fitting a multitude of intercommunicating miniature wireless sensor network devices within the cross section of the probe. Each of these sensing elements integrates an electrode to measure the electrical conductivity of the process flow using the Electrical Impedance Tomography (EIT) technique. Other sensors such as the pressure, capacitance, and ultrasonic sensors might be added in the future to provide more complementary information on the electrical and acoustic properties of the process fluid. The digitalized data is then transferred to an external remote transmitter which performs in real-time the two dimensional reconstruction of the image of the flow passing through the sensors array. This transmission of data is done in a time multiplexing manner, avoiding data collision, where each sensing node is allocated a predefined time slot. The cross sectional tomography images of static images which are reconstructed offline using this concept are presented.

ABSTRACT Computer simulations of two-phase gas-liquid pipe flow cases have been performed with the commercial code STAR-CD using the VOF method and the results are illustrated in this paper. Computed pressure losses for stratified flow and stratified-wavy flow and time-averaged liquid levels are compared against measured data, showing good agreement with the experimental data. To achieve these results, turbulence in the region close to the interface required the introduction of a law-of-the-wall based correction term in the turbulent energy dissipation equation. Two approaches for the turbulence treatment at the gas-liquid interface are briefly exposed and a comparison of the results is presented.

ABSTRACT In a recent paper, Gokcal et al. (2008) investigated the effects of viscosity on drift velocity in a horizontal pipe experimentally and theoretically. Experiments were conducted in a 50.8-mm (2-in.) ID horizontal pipe. A mechanistic model was developed. The results revealed that viscosity is a significant parameter for drift velocity which used to be a neglected factor in all previous literature. In this paper the effect of high viscosity in different pipe diameters on drift velocity is investigated using a Computational Fluid Dynamics (CFD) approach. New drift velocity data for 3-inch (76.2-mm) and 6-inch (152.4-mm) ID. pipes are presented. Experimental and simulation results are compared, and discussions on the effect of both diameter and liquid viscosity on the drift velocity for horizontal flow are made. The results of this work may improve the existing two-phase slug flow models for heavy oil.

ABSTRACT There are many sources of uncertainty in multiphase fluid flow models. However, most models are deterministic. One of the sources of uncertainty is closure relationships. In this paper, we present a wrap-around uncertainty propagation tool designed to study and quantify the impact of closure relationship uncertainties on multiphase flow model predictions. As a case study, we examined the impact of slug length uncertainty on pressure drop and liquid holdup predictions of Tulsa University Fluid Flow Projects (TUFFP) Unified Model(1) for two-phase vertical flows. The results suggest that a ± 50% slug length uncertainty can result up to 40% uncertainty in pressure drop predictions for some operating conditions, and almost no uncertainty for others.