Air-water Two-phase Flow Patterns Research Articles

Unraveling air–water two-phase flow patterns in water pipelines based on multiple signals and convolutional neural networks

Abstract Flow pattern identification (FPI) is crucial for evaluating air entrapment in water pipelines and ensuring the safety of pipeline operations. The presence of two-phase flow in water pipelines not only leads to pressure fluctuations but also induces pipeline vibration. However, current research has primarily focused on using pressure-related signals for FPI, and the analysis of vibration signals in FPI is rare. In this study, FPI in water pipelines is investigated based on convolutional neural networks (CNNs) using high-frequency vibration signals. The information fusion of vibration signals in FPI is newly proposed via the stacked generalization technique. The proposed method is compared with pressure signal-based FPI methods and the effect of signal sampling parameters on FPI accuracy is discussed. The results show that the performance of vibration signals (including axial or radial acceleration signals) outperforms pressure signals in both time and frequency domains. Moreover, the fusion of vibration signals shows the superior results compared to any univariate signals. The duration of sampling has a more significant impact on the results of FPI than the sampling frequency. This study provides a new way that FPI theory is applied to solve air entrapment evaluation in water pipelines.

Experimental investigation and one-way coupled fluid-structure interaction analysis of gas-liquid two-phase flow conveyed by a subsea rigid M-shaped jumper

The subsea rigid M-shaped jumpers provide reliable and flexible connections in subsea production systems by using deflection to reduce the impact of internal flow and providing installation tolerances. During their service, the gas-liquid mixed multiphase flow conveyed by the jumpers will cause flow-induced vibration (FIV) and reduce their fatigue life. In this paper, the air-water mixed multiphase flow conveyed by a rigid M-shaped jumper at different mixture velocities (1 m/s to 4 m/s) and volume fraction of water (0.2–0.8) is investigated to analyze the factors affecting the flow patterns and identify the most dangerous regions. Flow patterns in each segment and historical pressure changes are obtained by simulation and experiment. The dynamic response of the jumper excited by FIV is obtained by one-way coupled fluid-structure interaction (FSI). The evolution of the air-water two-phase flow pattern from inlet to outlet is visualized and analyzed. In particular, the regularity of Taylor bubble generation in the first ascending segment is discussed. The pressure tends to decrease from inlet to outlet, and a consistent change in pressure difference is observed between simulation and experiment. The elbows are the most notable segments, with the maximum deformation and stress concentration.

Unified two-component interfacial area concentration model for narrow channel two-phase flow

The conventional interfacial area concentration (IAC) models do not distinguish narrow and wide side interfaces when applied in narrow channels, which hinders accurate modeling of interfacial transfer mechanisms at different interfaces. In order to overcome such deficiency, void fraction data obtained in narrow-channel air-water flow with the flexible printed circuit (FPC) wire-mesh sensor (WMS) is analyzed to establish a two-component IAC model which consists of two essential parameters, i.e. the projected interfacial area concentration (PIAC) and projected interfacial length concentration (PILC). Based on the refined and sharpened void fraction distribution, the postprocessing algorithm is developed to obtain the PIAC and PILC for different flow patterns. Analysis of experimental data indicates that the PIAC is approximately proportional to void fraction, while variation of the PILC with void fraction is complicated. The ratio of PILC to PIAC, which characterizes topology of interfaces, shows distinct value in bubbly flow compared with other flow patterns. Such difference is attributed to the existence of large bubbles, typically with spanwise length over 70% of the channel width, in flow patterns other than bubbly flow. Hence, bubbles are separated into two groups with the critical ratio of bubble width to channel width in the model derivation. The semi-empirical PIAC and PILC models are derived for each bubble group. Based on volume fraction of two bubble groups, the unified PIAC and PILC models are obtained to realize the accurate prediction in all the flow patterns. The unified two-component IAC model and other IAC models are evaluated with the present IAC experimental data and the experimental data of Shen et al. (2012). The evaluation shows the superiority of the unified two-component IAC model and that the proposed unified two-component model can be applied in the narrow-channel air-water two-phase flow patterns with α¯≤0.9.

3D gas-liquid interfaces and flow characteristics of two-phase flows in horizontal tubes

In order to investigate the characteristics of gas-liquid two-phase flows in horizontal mini circular tubes with inner diameters of 3.14 and 6.68 mm, a prism is adopted to improve the light path in the visualization experimental setup. The front and top views of air-water two-phase flow patterns in two tubes are captured synchronously based on the improved method. Three-dimensional gas-liquid interfaces, flow pattern maps, and void fraction are obtained. The experimental results show that tube diameters have significant effects on flow patterns transition lines in the flow pattern maps, but the void fractions are independent on channel sizes. The effect of gravity gradually decreases with decreasing tube diameter, while that of surface tension is enhanced. As a consequence, the proportion of annular flow in flow pattern map increases in mini tubes, while the reverse is true for the stratified flow whose proportion decreases dramatically in mini channels. The void fraction increases with increasing gas quality. Experimental void fractions obtained using the three-dimensional gas-liquid interfaces fit well with correlations in the open literature. The shape of PDF distributions varies with flow patterns, which could be used to identify flow patterns in industrial applications.

Identifying Flow Patterns in Water Pipelines Using Complex Network Theory

AbstractAir pockets trapped in water pipelines are a common phenomenon and can lead to different air-water two-phase flow patterns: stratified, blowback, plug, and bubbly flows. The two former flow...

Upward two-phase flow patterns in vertical circular pipe under rolling condition

The experiments on the air-water two-phase flow patterns in a 24 mm diameter vertical circular pipe under both static and rolling conditions were conducted. The superficial velocities were in the range of 0.05–4.07 m/s and 0.02–20.55 m/s for water and air, respectively. Five typical flow pattern morphologies resulting from a rolling motion were observed: dispersed bubbles; bubbles; slug; churn; and annular flow. Flow pattern transitions were identified by analysis of visual images and a bubble chord-length probability distribution function (PDF). The static experimental data was used to validate existing flow pattern transition models that were then modified to fit rolling conditions by considering the additional forces caused by rolling motion. These models also agreed well with the experimental data. Investigation of the influence of the rolling motion on flow pattern transition showed that, as the rolling amplitude increases, the formation of dispersed bubble flow requires a higher superficial liquid velocity. It was also found that, the greater the rolling amplitude, the larger region of churn. The rolling period, however, has little impact on flow pattern transitions.

Surface hydrophobicity change on adiabatic two-phase flow pattern transitions in horizontal tubes

The effect of surface hydrophobicity on the air-water two-phase flow pattern transition was experimentally investigated. Experiments and the two-phase flow visualization were carried out in a hydrophobic polymer tube (4.76 mm ID). The tube material is fluorinated ethylene propylene (FEP) with a hydrophobic surface, and the static contact angle of a water droplet on a smooth surface is 97.28°. The channel surface was roughened to increase the hydrophobicity. The increased surface roughness results in a microstructured surface texture that forms a fine air trapped layer below the liquid droplet on the polymer surface. The channel surface was rotary ground by sandpaper number grit 400, 320, 100, and 80 to develop microstructure textured surface. The measured static contact angle of the roughened surface was 110.7°, 113.26°, 121.03°, and, 126.28° in order of increasing surface roughness. The two-phase flow of the channels with different contact angles was visualized by a high-speed camera and the flow pattern maps were constructed. Even under the same phase superficial velocity conditions, differences in the degree of hydrophobicity lead to transitions to entirely different patterns. The two-phase flow pattern map with the change in hydrophobicity is partitioned by the change in the modified Weber number as a factor indicative of the degree of surface hydrophobicity. The modified Weber number is defined as the ratio of flow inertia to adhesion force, and the new pattern map of the two-phase flow is redefined to take into account the wall hydrophobicity change.

Experimental study of air-water two-phase flow pattern evolution in a mini tube: Influence of tube orientation with respect to gravity

The evolution of two-phase flow patterns is investigated experimentally in a 1500 mm long capillary tube of 3 mm inner diameter with different orientations with respect to gravity. The flow patterns are recorded with a fast camera (1000 fps) at two locations along the tube under the same operating conditions. The first is close to the inlet zone at L/D = 10 and the second is far downstream at L/D = 420. Co-flowing configurations downward and upward are considered where air is injected through a nozzle of 110 μm of inner diameter and centered on the tube axis. Unlike many existing studies in the literature, the present work explores ranges of small superficial velocities of the two fluids, from 0.78 × 10−3 m/s to 79 × 10−3 m/s for the liquid phase and from 2.3 × 10−3 m/s to 3.54 m/s for the gas phase, in order to deal with smooth interfaces. The experiments show that the flow patterns are very sensitive to inlet conditions. Although, the superficial velocities considered are very small, several two-phase flow patterns are visualized namely bubbly flow, slug flow, unstable annular and annular flow. In the upward flow, the annular pattern is unstable, while for the downward flow, the bubbly pattern is difficult to obtain. Phenomena like coalescence and breakup yield to the evolution of the flow pattern from the entrance of the tube to the downstream location. The visualizations highlight also that certain flow patterns evolve from one structure to another along the tube from the location L/D = 10 to L/D = 420. Furthermore, the two-phase flow maps are drawn and the effect of gravity on the flow maps as well as on the flow pattern transitions is discussed.

Experimental Study on Hydraulic Characteristics of Pneumatic Lifting Pump

A pneumatic lifting pump is used in sewage treatment, offshore oil production, and other fields because of its simple structure and strong practicability. In order to study its internal hydraulic characteristics and gas-liquid two-phase flow, this paper carries out experimental research on the influence of different air intake modes and riser diameters on the performance of a pneumatic lifting pump. The air-water two-phase flow pattern in the riser and motion characteristics of bubble formation at the nozzle are obtained by a high-speed camera. Through theoretical analysis, the theoretical model of a pneumatic lifting pump is established, and experimental results verify the theoretical model well. The results show that when the submergence ratio is constant, the lifting efficiency decreases with the smaller intake area under different intake areas; and the influences of different holes distributions on liquid flow rate and lifting efficiency are not significant under the same intake area. At the same submergence ratio, the smaller the riser diameter, the smaller the final lifting liquid flow rate and the larger the lifting efficiency peak value.

Numerical study of air-water two-phase flow in a two-dimensional vertical helical channel

In this study, air-water two-phase flow patterns in a two-dimensional vertical helical channel are investigated. The simulations are performed using OpenFOAM software. The single-fluid method and Eulerian−Eulerian approach are used to solve the governing equations. The paper focuses on different flow patterns based on the change in the apparent velocity of water and air, and the influence of volume fraction of the air at the entrance of the channel. The variations of turbulent kinetic energy, volume fraction and velocity profile are presented. The range of air and water apparent velocities at the channel entrance is 0.05–10 m s−1 and 0.1–1.5 m s−1, respectively. According to the results, four flow patterns including bubbly, churn, turbulent, and annular are observed. It is demonstrated that for the bubbly regime, the values of volume fraction of phases are close to unity. When the air inlet velocity is very high, the two-phase flow forms the annular flow pattern.

Visualization Study of Air-Water Two-Phase Flow Pattern in Upward Vertical Flow of Square Channel

An important two-phase flow experiment study is carried out as a consideration in the application of piping system design that depends on two-phase flow. The purpose of this study was to determine the air-water flow pattern formed in a small square channel with vertical flow direction and to know the flow configuration. Channel 5 mm hydraulic diameter was used in this study. Bubble, slug, churn, and annular flow patterns are formed in the air superficial velocity range 0.33 - 7.31 m/s and superficial velocity of water is 0.13 - 1.2 m/s. A flow pattern map produced has not similarities with the mini-channel flow pattern map in previous studies. The velocity of bubbles and void fractions increases with the increase of superficial air velocity. The void fraction of present research has a good correlation to the results of predictions proposed by previous studies.

Air injection methods: The key to a better performance of airlift pumps

This paper presents a comprehensive experimental investigation on the effect of air injection methods on the performance of airlift pumps. The effect of air–water two-phase flow patterns and the phase interaction is presented. The present airlift pump utilizes various air injectors specially designed to enable air injection radially, axially, dual (radial and axial), or with a swirl effect at the air inlet ports. In addition, two air injection modes including steadily and pulsating, at different injection frequencies, are investigated at different two-phase flow patterns. The results have shown that the performance of airlift pumps does not only depend on the design of air injector, but also on the dynamics of air injection method. The pulsating mode of air injection is found to control the intermittent flow pattern in the pump riser and consequently able to keep the pump operating within the best efficiency range. The maximum efficiency of the pulsating axial injection operating at a frequency of 1Hz has improved by 60% over that of typical steady axial injection and by 24% over that of steady dual injection. Furthermore, the performance of airlift pumps with dual injector is improved by synchronizing the injected air at both axial and radial injection ports. The improvement in this arrangement is attributed to the better utilization of available compressed air, reduction of axial air penetration and minimization of water reverse flow in the riser pipe.

Gas–Liquid Two-Phase Flow in a Mini-Channel with Water Infiltration Continuously Along the Flow Direction

Characteristics of air–water two-phase flow pattern in a square-mini-channel were investigated experimentally. One sidewall of the flow channel was made of a permeable porous media. Air was fed into the channel from its inlet, while liquid water was injected uniformly into the channel along the permeable sidewall. Based on the results of visualization, the typical flow patterns encountered in the conventional co-current gas–liquid two-phase flow were not observed in the present work. Flow patterns of this special gas–liquid two-phase flow were catalogued, and the influence of gas flow rate and liquid infiltration rate on flow patterns was also analyzed. The results also indicate that the channel differential pressure on both sides brings obvious fluctuations when the liquid is infiltrated. The frequency of this fluctuation depends not only on the amount of liquid infiltration but also relies on the gas flow rate significantly. However, the amount of liquid infiltration barely affects the fluctuation in the amplitude of the pressure differential.

Experimental study of two-phase flow pattern evolution in a horizontal circular tube of small diameter in laminar flow conditions

The carried out experimental study aims at analyzing air–water two-phase flow patterns in a horizontal tube for laminar flow with carefully controlled conditions. An experimental test rig has been setup to produce a two-phase flow in a glass tube of 3mm inner diameter with co-current air and water flows. Unlike previous studies, the investigated ranges of superficial velocity of the liquid and gas are from 0.78×10−3ms−1 to 79×10−3ms−1 and from 2.3×10−3ms−1 to 3.54ms−1 respectively. Flow visualizations are recorded with a high-speed camera in the entrance region at L/D=10 corresponding to a mixing zone of the two phases and far downstream at L/D=420. The results are used to highlight the flow patterns in both zones. Within the explored ranges of superficial velocities, several flow patterns were obtained namely bubbly, slug and annular flow pattern. More complex flow patterns were visualized in the entrance region that evolve to fewer and simplified flow structure with smoother interfaces far downstream. Flow maps are drawn and proposed. They reveal some pattern rearrangements. A detailed description and the physical interpretation of the change in the flow structure between the two investigated zones are given in the paper.

Generation of Air–Water Two-Phase Flow Patterns by Altering the Helix Angle in Triple Helical Microchannels

Flow of a liquid inside a helical tube is composed of axial and circumferential components, the latter arising because of its specific geometry. For a gas–liquid two-phase flow inside a helical tube, the coupled effect of these two flow components leads to a variety of flow patterns, for example, slug, bubble, and stratified flow. We present here a novel triple-helical microchannel, in which, the two-phase flow is found to engender several additional flow patterns not observed with the conventional geometries, for example, the parallel and oscillating annular flow and even simultaneous occurrence of several such patterns. We show that the transition between these patterns depends not only on the fluid rates of the two liquids but also on the helix angle. We have presented detailed phase diagrams to elaborate these effects. We have examined also the effect of channel geometry on the specific features of these flow patterns.

An Experimental Study on Air-Water Two-Phase Flow Patterns in Pebble Beds

An experimental study was conducted on the air-water two-phase flow patterns in the bed of rectangular cross sections containing spheres of regular distribution. Three kinds of glass spheres with different diameters (3 mm, 6 mm, and 8 mm) were used for the establishment of the test section. By means of visual observations of the two-phase flow through the test section, it was discovered that five different flow patterns occurred within the experimental parameter ranges, namely, bubbly flow, bubbly-slug flow, slug flow, slug-annular flow, and annular flow. A correlation for the bubble and slug diameter in the packed beds was proposed, which was an extended expression of the Tung/Dhir model, Jamialahmadi’s model, and Schmidt’s model. Three correlations were proposed to calculate the void friction of the flow pattern transition in bubble flow, slug flow, and annular flow based on the bubble model in the pore region. The experimental result showed that the modified Tung and Dhir model of the flow pattern transition was in better agreement with the experimental data compared with Tung and Dhir’s model.

Two-phase Flow Patterns in a Four by Four Rod Bundle

Air-water two-phase flow patterns in a four by four square lattice rod bundle consisting of an acrylic channel box of 68mm in width and transparent rods of 12mm in diameter were observed by utilizing a high speed video camera, FEP (fluorinated ethylene propylene) tubes for rods, and a fiberscope inserted in a rod. The FEP possesses the same refractive index as water, and thereby, whole flow patterns in the bundle and local flow patterns in subchannels were successfully visualized with little optical distortion. The ranges of gas and liquid volume fluxes, {JG} and {JL}, in the present experiments were 0:1 ≪ {JL} < 2:0 m/s and 0:06 < {JG} < 8:85 m/s, which covered typical two-phase flow patterns appearing in a fuel bundle of a boiling water nuclear reactor. As a result, the following conclusions were obtained: (1) the region of slug flow in the {JG}-{JL} flow pattern diagram is so narrow that it can be regarded as a boundary between bubbly and churn flows, (2) the boundary between bubbly and churn flows is close to the boundary between bubbly and slug flows of the Mishima and Ishii's flow pattern transition model, and (3) the boundary between churn and annular flows is close to the Mishima and Ishii's model.

Two-phase Flow Patterns in a Four by Four Rod Bundle

Air-water two-phase flow patterns in a four by four square lattice rod bundle consisting of an acrylic channel box of 68mm in width and transparent rods of 12mm in diameter were observed by utilizing a high speed video camera, FEP (fluorinated ethylene propylene) tubes for rods, and a fiberscope inserted in a rod. The FEP possesses the same refractive index as water, and thereby, whole flow patterns in the bundle and local flow patterns in subchannels were successfully visualized with little optical distortion. The ranges of gas and liquid volume fluxes, {JG} and {JL}, in the present experiments were 0:1 ≪ {JL} < 2:0 m/s and 0:06 < {JG} < 8:85 m/s, which covered typical two-phase flow patterns appearing in a fuel bundle of a boiling water nuclear reactor. As a result, the following conclusions were obtained: (1) the region of slug flow in the {JG}-{JL} flow pattern diagram is so narrow that it can be regarded as a boundary between bubbly and churn flows, (2) the boundary between bubbly and churn flows is close to the boundary between bubbly and slug flows of the Mishima and Ishii's flow pattern transition model, and (3) the boundary between churn and annular flows is close to the Mishima and Ishii's model.

Gas–liquid two-phase flow patterns in a miniature square channel with a gas permeable sidewall

Characteristics of air–water two-phase flow patterns in a miniature square channel having a gas permeable sidewall were investigated experimentally. Water was fed into the channel from its entrance, while air was injected uniformly into the channel along the permeable sidewall. This configured two-phase flow problem is encountered in direct feed methanol fuel cells. Flow patterns in both vertical upward and horizontal flows were identified using a high-speed motion analyzer. The visualization shows that the typical flow pattern encountered in the conventional co-current gas–liquid two-phase flow, such as bubbly flow, plug flow, slug flow and annular flow were also observed in the present work. However, unlike the conventional co-current gas–liquid two-phase flow in a channel with gas and liquid uniformly entering from one of its ends, for the flow configuration considered in this work, the stratified flow and wavy flow were not found in horizontal flow. And a so-called “single layer bubbly flow” was found in vertical upward flow, which is characterized by a mono small-gas-bubble layer existing adjacent to the surface of the permeable sidewall with the reminding space occupied by the liquid phase. Four transitional flow patterns such as bubbly-plug flow, bubbly-slug flow, plug–slug flow, and slug-annular flow, were found to exist between the distinct flow patterns. Finally, the flow regime maps for various liquid volumetric fluxes are presented in terms of mass quality versus the volumetric flux of gas phase.

Co-current air–water two-phase flow patterns in vertical triangular microchannels

Characteristics of co-current upward air–water two-phase flow patterns in vertical equilateral triangular channels with hydraulic diameters of 2.886, 1.443 and 0.866 mm were investigated experimentally. Flow regimes were identified by both visual observations using a high-speed motion analyzer and dynamic pressure-drop measurements. The experimental results show that the typical flow patterns encountered in the conventional, large-sized vertical circular tubes, such as dispersed bubbly flow, slug flow, churn flow and annular flow, were also observed in the channels having larger hydraulic diameters ( d h=2.886 and 1.443 mm). However, for the smallest channel ( d h=0.866 mm), dispersed bubbly flow pattern, characterized by randomly dispersed bubbles in continuous liquid phase, was not found, although the other typical flow patterns remained in the channel. Moreover, the experiments reveal that, for the channel of d h=0.866 mm, a so-called capillary bubbly flow pattern, characterized by a single train of bubbles, essentially ellipsoidal in shape and spanning almost the entire cross-section of the channel, existed at low gas flow rates. It is further found that in the slug flow regime, slug-bubbles were substantially elongated. Finally, flow regime transition boundaries for the triangular microchannels were compared with relevant flow regime transition models and correlations as well as the existing experimental data for small round tubes and square channels.