Search

A thorough understanding of the behaviour of the transition from plug to slug flow is imperative based on the fact that the transition can trigger an abrupt radial pressure variation. This can bring about major vibrations to the pipeline and also lead to big differences in pressure gradient and wall temperature. Unfortunately, the transition from plug to slug flow is poorly understood due to the scarce availability of experimental data. Furthermore, a considerable amount of research on the transition from plug to slug flow in the literature is based on an air–water system. In this work, the transition behaviour from plug to slug flow in a horizontal pipeline was experimentally investigated using electrical capacitance tomography (ECT). Working fluids are air–silicone oil. The work was carried out over a range of liquid and gas superficial velocities of 0.05–0.47 m/s and 0.05–4.7 m/s, respectively. Wave growth and wave characteristics mechanisms were observed to be responsible for the transition from plug to slug flow based on the obtained experimental results. Both liquid and gas superficial velocities have a major influence also on these mechanisms. The drift flux parameters for the transition from plug to slug flow was determined. A reasonably good agreement was observed from the comparison between present experimental data against hitherto published empirical models and correlations.

The increasing use of commercial flow simulators in the oil and gas industry provides the impetus to accurately model multiphase flows in pipelines. The current work investigates stratified oil-water flows in horizontal pipelines by experimentally measuring interfacial deformations, correlating them to interfacial friction. The derived interfacial model is then implemented into the two-fluid model to predict pressure losses in pipelines. An experimental campaign was undertaken using process oil ( ρ = 845 k g / m 3 , μ = 0.03 P a . s ) and water as the test fluids in acrylic pipes of 40 mm diameter, with objectives of capturing interfacial deformations via image processing techniques, while also measuring pressure gradients in the pipe. The interfacial deformations were seen to increase with increasing flow velocities, plateauing with wave amplitudes of 3 mm. Further tests with identical fluids, conducted on larger diameter pipes (54.8 mm and 108.4 mm) also showed similar trend. Wave aspect ratios were correlated with a modified Froude number, linking the interfacial deformations to input flow variables. An empirical equation, utilizing the amplitude of interfacial deformations as roughness heights is derived for the interfacial friction factor. Incorporating the derived equation into the one dimensional two-fluid model, pressure losses were calculated and compared against experimental pressure drop measurements conducted in current work, as well as with those in literature. The performance of this model is also compared with existing prediction models and is found to provide improved accuracies across wide range of viscosity ratios and pipe diameters.

Annular flow is predominant in gas wells. Liquid may be present in form of entrained droplets as well as in the liquid film on pipe wall. The knowledge of the average liquid film thickness is vital for detailed mechanistic modelling of churn–annular flow in engineering applications. So far, the models for liquid film thickness prediction are limited to vertical and horizontal pipes. These models were based on limited ranges of experimental data. In addition, exhaustive iterations are needed when these models are used to estimate liquid film thickness. In this work, a new correlation to predict liquid film thickness in a 180o bend under gas–liquid annular conditions was successfully proposed. The correlation was based on dimensionless numbers (modified gas and liquid Weber numbers and gas Froude number) which reflect the underlined physics governing gas–liquid interaction in the bend. The Weber numbers capture the two important forces (inertial and surface tension forces) which govern the formation of the liquid film thickness within the system while gas Froude number quantifies the interaction of two important forces (centrifugal and gravitational forces) which determines the distribution of the phases across a bend. The proposed liquid film thickness correlation was based on the experimental data obtained from a wide range of operating conditions. The liquid superficial velocities ranges from 0.02 to 0.2 m/s and gas superficial velocities from 3.5 to 16 m/s at different measurement locations of 45o, 90o and 135o of the bend with a diameter of 127 mm. The liquid film thickness in air–water and helium–water annular flow can be predicted by . The validation of the proposed correlation used to predict liquid film thickness in gas–liquid annular systems with different pipe diameters were examined against the available data in the literature. Good agreement was found between the predicted values of liquid film thickness with the experimental data at different measuring locations of the bend.

The characterization of two-phase flow in both vertical and horizontal pipes is important in oil/gas transportation. As the pipe orientation may change from well to well, it is also important to understand the characteristic responses expected of a two-phase mixture to a change in pipe inclination. This study concerns the changes in the void fraction, structure frequency and structure velocity that occur within an air–silicone oil mixture as a function of the pipe orientation. Experimental data were obtained from the combined use of electrical capacitance tomography (ECT) and wire mesh sensors (WMS), which allow the 3D visualization of the flow patterns. The reported experiments were performed on a 67 mm diameter pipe, in a flow loop, in which a pipe section may be inclined at angles of between −5 or 90° to the horizontal. For this study, the inclined pipe was either set at an angle of 0 or 90 to the horizontal, which correspond to a horizontal or vertical pipe setting, respectively. The results of flow development using the PDF of void fraction obtained at 3 measurement locations; ECT 1, ECT 2 and WMS at 4.4, 4.489 and 4.92 m, respectively, showed that the flow is fully developed and statistically stable for the vertical two-phase flow. While on the other hand, the flow is in rapid development for the horizontal two-phase flow scenario at same liquid and gas superficial velocities. The processed data reveal the differences in flow distribution produced by the pipe inclination. Within the vertical pipe configuration spherical cap bubbles, slug and churn two phase flow patterns were observed, whilst plug, slug and stratified wavy two-phase flows were identified in the horizontal configuration of the pipe. It is concluded that a plot of mixture superficial velocity against average void fraction may be used to provide a qualitative assessment of the flow patterns observed, regardless of pipe inclination. The two-phase flow variables that are concluded to significantly influence the distribution coefficient, C0, and drift velocity, VD, are the pipe orientation and flow patterns.

Reliable and accurate prediction of the transition from spherical cap bubble to slug flow is crucial not only to the operation of industrial facilities such as the crude oil pipelines, bubble column, and nuclear reactors but also for model development for computational fluid dynamics (CFD) studies. The present paper presents a review of the transition mechanics from spherical cap bubble flow to slug flow in vertical pipes. The bubble flow was split into sub-regions, bubbles and spherical cap bubbles and the mechanisms to classify them (i.e., bubble terminal velocity and cap bubble velocity) was analysed. For now, the literature review does not present some important previous works. This paper presents an original data set of gas–silicone oil in vertical pipes to support the new findings. The experimental two-phase data classifies the flow patterns using the probability density function (PDF) and shows the important flow variables such as average void fraction, pressure gradient, slug body void fraction, liquid slug, Taylor bubble and slug unit lengths, structural velocity and frequency obtained by electrical capacitance tomography (ECT) and a wire mesh sensor (WMS).

This paper investigates the unsteady hydrodynamic behaviour of slug flow occurring within an air–silicone oil mixture, within a horizontal 67 mm internal diameter pipe. A series of slug flow regime experiments were performed for a range of injected air superficial velocities (0.29–1.4 m s−1) and for liquid flows with superficial velocities of between 0.05–0.47 m s−1. A pair of Electrical Capacitance Tomography (ECT) probes was used to determine: the slug translational velocities of the elongated bubbles and liquid slugs, the slug frequencies, the lengths of elongated bubbles and the liquid slugs, the void fractions within the elongated bubbles and liquid slugs. The pressure drop experienced along the pipe was measured using a differential pressure transducer cell (DP cell). A comparative analysis of the current experimental data and that previously published experimental confirms good agreement.

This paper looks into the results of an experimental study concerned with the phase distributions of gas–liquid multiphase flows experienced in a vertical riser. Scale experiments were carried out using a mixture of air and silicone oil in a 6 m long riser pipe with an internal diameter pipe of 67 mm. A series of pipe flow experiments were performed for a range of injected air superficial velocities over the range 0.05–4.73 m/s, whilst the liquid superficial velocities ranged from 0.05 to 0.38 m/s. Measurements of cross-sectional void fraction and radial time averaged void fraction across a pipe section located 4.92 m from the pipe flow injection were obtained using a capacitance wire mesh sensor (WMS). The data were recorded at a frequency of 1000 Hz over an interval of 60 s. For the range of flow conditions studied, the average void fraction was observed to vary between 0.1 and 0.83. An analysis of the data collected concluded that the observed void fraction was strongly affected by the gas superficial velocity, whereby the higher the gas superficial velocity, the higher was the observed average void fraction. The average void fraction distributions observed were in good agreement with the results obtained by other researchers. The accuracy and performance of void fraction correlations were carried out in terms of percentage error and Root Mean Square (RMS) error. Reasonably symmetric radial void fraction profiles were obtained when the air–silicone oil was fully developed, and the shape of the symmetry profile was strongly dependent on the gas superficial velocity. The data for air/water and air/silicone oil systems showed reasonably good agreement except at gas superficial velocity of 0.05 m/s. A comparison of the experimental data was performed against a published model to investigate the flow structure of air–water mixtures in a bubble column. A satisfactory report was observed for radial void fraction profile (mean relative error is within 5.7%) at the higher gas superficial velocities.

This paper presents the results of comparison of experimental and CFD studies of slug flow in a vertical 90° bend using validated models. For the experimental part, Electrical Capacitance Tomography (ECT), Wire Mesh Tomography (WMS), and high-speed videos were used to monitor an air-silicone oil mixture flowing in a vertical 90o bend. The ECT probes were mounted before the bend whilst the WMS was positioned either immediately upstream or immediately downstream of the bend. The downstream pipe was maintained horizontal whilst the upstream pipe was maintained vertical. The bend (R/D = 2.3) was made of transparent acrylic resin. Parallel to the experiments, simulations were carried out for same experiment set-up using the commercial software package Star-CD and Star-CCM+. The condition was simulated with the Volume of Fluid (VOF) model. The simulation predictions were validated against the experimental data. A reasonably good agreement was observed for the results of the experiment and CFD.

The grid generation issues found in the 3D simulation of two-phase flow in a pipe using Computational Fluid Dynamics (CFD) are discussed in this paper. Special attention is given to the effect of the element type and structure of the mesh. The simulations were carried out using the commercial software package STAR-CCM+, which is designed for numerical simulation of continuum mechanics problems. The model consisted of a cylindrical vertical pipe. Different mesh structures were employed in the computational domain. The condition of two-phase flow was simulated with the Volume of Fluid (VOF) model, taking into consideration turbulence effects using the k-e model. The results showed that there is a strong dependency of the flow behaviour on the mesh employed. The best result was obtained with the grid known as butterfly grid, while the cylindrical mesh produced misleading results. The simulation was validated against experimental results.

A conductance wire mesh sensor system has been employed on a vertical 67 mm diameter pipe with the up flow of air and water mixtures. The measuring system provides time and cross-sectionally resolved information about the spatial distribution of the phases. Statistical information can be extracted and used to identify flow patterns. The fully resolved data has revealed a hitherto unreported structure has been seen in churn flow which could be linked to the wisps in wispy-annular flow.

A wire-mesh sensor has been employed to study air/water and air/silicone oil two-phase flow in a vertical pipe of 67 mm diameter and 6 m length. The sensor was operated with a conductivity-measuring electronics for air/water flow and a permittivity-measuring one for air/silicone oil flow. The experimental setup enabled a direct comparison of both two-phase flow types for the given pipe geometry and volumetric flow rates of the flow constituents. The data have been interrogated at a number of levels. The time series of cross-sectionally averaged void fraction was used to determine characteristics in amplitude and frequency space. In a more three-dimensional examination, radial gas volume fraction profiles and bubble size distributions were processed from the wire-mesh sensor data and compared for both flow types. Information from time series and bubble size distribution data was used to identify flow patterns for each of the flow rates studied.

This paper presents a comparison of the results obtained from experiments and CFD studies of slug flow in a vertical riser. A series of two experimental investigations were carried out on a 6 m vertical pipe with a 0.067 m internal diameter charged with an air–silicone oil mixture. For the first set of experiments, the riser was initially full of air, and then liquid and gas flows set to liquid and gas superficial velocities = 0.05 and 0.344 m/s, respectively, electrical capacitance tomography (ECT) and wire mesh sensor (WMS) transducers were employed. In the second one, the riser was initially full of (static) liquid, and then liquid and gas flows set to liquid and gas superficial velocities = 0.05 and 0.344 m/s, respectively, only ECT was used. A characterisation of the observed slug flow regimes was carried out. This includes the evaluation of the instantaneous distribution of the phases over the pipe cross-section, the Probability Density Function (PDF) of void fraction, time series of cross-sectional void fraction, Power Spectral Density (PSD), structure velocity of the Taylor bubble, lengths of the liquid slug and Taylor bubble and void fractions in the liquid slug and Taylor bubble. The simulation results were validated both qualitatively and quantitatively against experimental data. A reasonably good agreement was observed between the results of the experiment and CFD.

Experiments have been carried out in a 67 mm diameter pipe with air and deionised water in which gas superficial velocities between 0.04 and 0.55 m/s were studied. A capacitance technique, employing electrodes mounted on either side of the outside of the acrylic resin pipe, has been used to provide time varying cross-sectionally averaged void fraction. Probability Density Functions of these data has permitted characterisation of the slug flow. Pressure drop measurements were made using differential pressure transducer. The mean void fraction, the void fraction in the slug region, the slug velocity and the dimensionless length of liquid slug all increase with the superficial gas velocity. On the other hand increasing the superficial gas velocity was shown to decrease the fractional slug length, the frequency and the film thickness around the Taylor bubble as well as the pressure gradient. A new version of the drift flux equation for bubble rise velocity is proposed and found to perform well against the present data and that from the literature.

This paper presents an investigation of the hydrodynamics of slug flow in a vertical 67 mm internal diameter riser. The slug flow regime was generated using a multiphase air–silicone oil mixture over a range of gas (0.42

The structure of gas liquid flow in horizontal and vertical pipes to some extent is well understood. However, the situation in inclined pipes is much more difficult with very little work published in literature. Changes in physical phenomena occur as the pipe inclination angle varies from the vertical through to inclined and then to horizontal. This work describes a study carried out at the University Of Nottingham on the effects of inclination on gas / liquid slug flow. Two advanced tomography techniques were applied simultaneously to the flow of a mixture of air and silicone oil in a 67 mm internal diameter pipe and the pipe was inclined at angles 0, 5, 10, 30, 45, 60, 80, 90 degrees. This paper reports on the use of twin plane electrical capacitance tomography (ECT) system developed by TomoFlow electronics Ltd to measure flow characteristics in gas-liquid flows. We report measurements over a range of liquid superficial velocities from 0.05 m/s to 0.5 m/s and gas superficial velocities from 0.06 m/s to 6 m/s at all the above angles in a pipe 6 m long. A second technique, Capacitance Wire Mesh Sensor (WMS) developed at Forschungszentrum Rossendorf-Dresden/Germany was also present in the tests, The results for the two sensors are shown to be within 1% of each other in some instances when comparing cross-sectional averaged void fraction. The data was recorded at an acquisition frequency of 1000 Hz over an interval of 60 seconds. This enabled an examination of the flow to be carried out at several levels of complexity. Both measuring sensors provide time and cross-sectionally resolved information about the spatial distribution of the phases. In present paper, the effect of inclination on the characteristics of slug flow is presented. Radial gas volume fraction profiles and bubble size distributions were also processed from the wire-mesh sensor output. The results indicate that the pipe inclination has a significant effect on the slug flow characteristics.Copyright © 2011 by ASME

Novel experimental data for phase separation of air-water mixtures in horizontal 90 deg branched elbows are presented in this work. The branched elbows were formed by attaching a pipe to a 90 deg elbow on the side of maximum radius of curvature, and halfway between the inlet and outlet sections of the elbow. All three arms coming from the junction were in the horizontal plane. Both the branch orientation and branch diameter were varied. Three different branch/elbow diameter ratios were tested, as well as three different branch inclination angles. In addition, the static pressure was monitored at different points along the ramified elbow using a set of pressure transducers in order to analyze and associate the pressure drop with the phase separation. At the inlet section of the elbow, the two-phase flow pattern was mainly slug flow. Based on the experimental data, a correlation for the liquid phase separation is proposed. Finally, the volume-weighted phase separation in the branched elbow was compared with the phase separations on the T-junction, and it was found that in some cases the branched elbows have a similar performance to that of the T-junctions.

Multiphase flows are common in industrial settings and bends in pipe lines cannot be avoided due to space limitations. Gas-liquid two phase flows could form material discontinuities that could have adverse effect on productivity and the pipe network due to sudden variations resulting due to the rapid momentum flux variations at fittings such as bends. Research into gas-liquid flow and bends can be motivated by the effect of the bend on the flow downstream of it which could alter the flow pattern occurring and the performance of downstream equipment. Alternatively, the interest might come from what occurs in the bend itself, there could be dryout of the film on the walls and consequent damage to the heat transfer equipment. Here we present measurements made with a number of accurate and fast responding sensors on three cases, two on the effect of the bend and one considering effects in the bend. The results show that the flow transformations occur in two phase flows depending on the orientation of the bend and the change could be captured using fast sweeping measurement techniques. We present the evidence of effectiveness of several types of measurement techniques that could fit into various combinations of phases. The results, point to how to achieve certain flow patterns. Also recommendations are provided regarding the position of any sensor installed to determine flow pattern.

This paper presents the results of a series of experiments conducted on a vertical 67 mm internal diameter riser to study the characteristics of slug flow generated within a multiphase air-silicone oil mixture, for a range of injected superficial gas and liquid flow rates. Air/ silicone oil flows in a riser have been investigated using advanced instrumentation called Electrical Capacitance Tomography (ECT), described in Hammer (1983), Huang (1995) and Zhu et al. (2003). The ECT was used to determine: the velocities of the Taylor bubbles and liquid slugs, the slug frequencies, lengths of Taylor bubbles and liquid slugs, the void fractions within the Taylor bubbles and the liquid slugs over a range of superficial liquid (0.047≤USL≤0.38 m/ s) and gas (0.344≤USG≤0.745 m/ s) velocities. An examination of the experimental data concluded that the rise velocity of the observed Taylor bubbles (structure velocity) were strongly dependent on the mixture superficial velocity. The void fraction within the liquid slug and the Taylor bubbles were observed to increase with gas superficial velocity. The lengths of liquid slug and Taylor bubbles were also observed to increase with an increase in the gas superficial velocity. The liquid slug frequency was observed to increase as the liquid superficial velocity increases, but was observed to be weakly dependant on the superficial gas velocity.

Knowledge of how the presence of a bend can change the flow patterns of a gas–liquid mixture is important for the design of multiphase flow systems, particularly to prevent burn-out and erosion–corrosion. Burn-out and erosion–corrosion both have serious implications for heat and mass transfer. The objective of this work therefore is to train an artificial neural network (ANN), a powerful interpolation technique, to predict the effect of a vertical 90o bend on an air–silicone oil mixture over a wide range of flow rates. Experimental data for training, validation, testing and final prediction were obtained using advanced instrumentation, wire mesh sensor (WMS) and high speed camera. The performance of the models were evaluated using the mean square error (MSE), average absolute relative error (MAE), Chi square test (X2) and cross correlation coefficients (R). The performance discriminator X2 for prediction of average void fraction is 2.57e-5 and that for probability density function (PDF) of void fraction MAE is 0.0028 for best performing models. The well trained ANN is then used to predict the effects of the two input parameters individually. The predicted results show that for the before the bend scenario, the most effective input parameter that reflects a change in flow pattern is the gas superficial velocity. On the other hand, the most unfavourable output parameter to measure after the bend is the average void fraction based on the fact that the flow near the bend is a developing one.