Gas-liquid annular flow at a vertical tee junction—Part I. Flow separation

Flow and heat transfer model for turbulent-laminar/turbulent gas-liquid annular flows

Fluid flow monitoring is essential in many industrial processes. However, some flow measurement methods may not be feasible in a few applications such as the ones in potentially explosive atmospheres or under presence of intense electromagnetic fields. In this study, we present the construction and evaluation of an optical fiber transducer, which is suitable for flow monitoring in industrial environments pipes. Besides single-phase flow, which has been already monitored by similar techniques in the past, we also present in this paper the measurement of two-phase gas-liquid flows. Working principle is based on the measurement of the force induced by the streaming fluid into a cantilever whose deformation is indirectly and optically monitored by means of fiber Bragg gratings. Static and dynamic tests were performed for transducer calibration and validation. In static calibration, sensor sensitivity to force corresponded to 2.34 nm/N. The transducer was successfully applied to measure single-phase flow rate in the range of 0 $\text{m}^{{3}}$ /h to 6.68 $\text{m}^{{3}}$ /h, presenting maximal uncertainty of 0.3 $\text{m}^{{3}}$ /h. Finally, the transducer was applied in two-phase gas-liquid flow, being able to correctly identify gaseous and liquid phases since the cantilever is exposed to different forces during the passage of the gas and liquid. In this way, the developed transducer may be well applied to two-phase flow monitoring.

Experiments of density wave instability in a sodium boiling two-phase flow in an annulus were carried out with the parameters of heat flux from 80 to 976kW/m2, inlet subcooling from 25.6 to 226.8°C, mass flow rate from 7.92 to 68.9 kg/h, and system pressure from 2,600 Pa to 0.06 MPa. It was found that the density wave instability occurred in the case of low exit quality, and the oscillation of flow rate was so large that the flow would be reversal. The lower inlet temperature, the higher system pressure and the larger mass flow rate could result in a more stable boiling two-phase flow. The oscillation period of the instability increased with the system pressure and the inlet subcooling, but it decreased with the mass flow rate. A correlation for the onset condition of the density wave instability was obtained from the experimental data.

Fines migration and pore clogging induced by single- and two-phase fluid flows in porous media: From the perspectives of particle detachment and particle-level forces

Pressure drop and liquid holdup in high pressure trickle-bed reactors

Two-phase cross flow over heat exchanger tubes creates vibrations which contribute greatly to the wear on the tubes. Fluidelastic instability is a major mechanism by which tubes can fail. In this work, the fluidelastic instability of a tube placed in an array subjected to two-phase cross flow has been studied. For the determination of fluidelastic instability, a triangular tube array was used. The tubes were made of acrylic and were 2.2 cm or 2.37 cm in diameter and 20 cm in length. Eighteen tubes and 4 half tubes formed 5 rows with a pitch to diameter ratio of 1.4. All of the tubes except the test tube were rigidly supported at the text section wall. The test tube was flexibly supported with two cantilever beams. By installing cantilever beams horizontally and vertically, drag and lift direction tube vibration were studied. Parameters of tube mass, structural stiffness, natural frequency, and pitch to diameter ratio were varied. The drag coefficients on a rigidly held tube in an array subjected to two-phase cross flow were measured. The tube in an array was located at displaced positions as well as at the normal position in order to study the variation of fluid force as the tube vibrates. In the experiments, gap Reynolds numbers up to 1 x 105 were obtained, while void fraction was varied from zero to 0.5. The drag coefficients in two-phase flow are much higher than those in single phase flow. The ratio of two-phase to single phase drag coefficient decreases as Reynolds number increases. The drag coefficient on a tube in an array increases as the tube is displaced in the direction of flow. The drag coefficient increases rapidly when the tube is displaced more than a certain critical distance.

Fluidelastic instability of flexible tubes subjected to two-phase internal flow

A model for droplet entrainment in heated annular flow

An overview of a measurement method for liquid film thickness in annular flows based on cold neutron imaging is given here. Neutron imaging being a non-intrusive, contactless method is attractive option for two-phase flow investigations offering an excellent contrast. It can provide with information at a high spatial resolution on the flow structure, like the thickness of the liquid film in annular flows. The method has been optimized, and its performance, regarding bias, statistical accuracy, upper and lower detection limits, has been thoroughly quantified using computational tools and measurement results. The technique has been developed based on nuclear fuel bundle models; however, it is applicable practically to annular flows in any arbitrary flow channel geometry of interest.

This study describes the concept of Liquid-Assisted Gas-Lift (LAGL). This concept includes three main parts: two-phase downward flow in annulus, two-phase flow through orifice Gas-Lift Valves (GLVs), and upward two-phase flow in pipes. The latter part is well described in the literature and will not be investigated in this work. However, there is a lack of studies on two-phase downward flow in annulus and through GLVs. Therefore, these two topics are investigated in this work. An experimental and numerical study on two-phase flow through orifice GLVs is presented. The experimental results are compared to predictions using a numerical model published in the literature for two-phase flow through restrictions. It is observed that the mechanistic model could successfully characterize two-phase flow thorough gas-lift valves with errors lower than 10%. A flow regime map of downward two-phase flow in the annulus is used to analyze the flow regimes observed during field-scale experiments of LAGL unloading. It was concluded that the intermittent and bubbly flow are preferred to minimize injection pressure during LAGL unloading operations. A commercial flow simulator is used to perform sensitivity analysis of the LAGL system and to perform unloading of a well using the LAGL concept. The unloading simulation results indicated that the use of LAGL has the potential to decrease the injection pressure requirement in more than 50% when compared to the conventional single-point gas injection. The simulation model is shown to be a useful tool to perform system analysis and optimization of LAGL.

Gas well productivity is affected by two distinct mechanisms: (1) liquid blocking, and (2) high velocity flow in two-phase. The former has been studied extensively recently but the understanding of the latter is limited. High velocity gas flow in single phase has been thoroughly studied by a large number of authors. Despite the fact that high-velocity coefficient in the presence of an immobile and a mobile liquid phase is much higher than that in single phase, only a handful of studies have been carried on the subject. In this work, we have measured the high-velocity coefficient, β in two-phase gas-liquid flow. The results are presented as a function of liquid relative permeability and also liquid saturation. In our measurements, the wetting state is varied by the treatment with a fluorochemical compound. Then the effect of wettability on the high-velocity coefficient in two-phase flow is investigated. Results show that when the liquid is strongly wetting, the high-velocity coefficient increases some 270-fold in water-gas two-phase flow. However, our data show a systematic reduction of high-velocity coefficients for the altered wetting state in two-phase flow. We present measurements of the velocity coefficients in single-phase flow, and in two-phase flow for both oil-gas and water-gas flow for strong liquidwetting and altered-wetting states. Based on our measurements, we conclude that the treatment of the wellbore region can result in significant improvement in well deliverability from the large reduction of high-velocity coefficients.

Single-phase and gas-liquid two-phase pressure drops caused by a sudden contraction in microtubes were experimentally investigated at room temperature and atmospheric pressure, using nitrogen and water. The experimental results on pressure drop with a novel measurement method, the tiny gaps on the tubes, were used to characterize the sudden contraction pressure drop for tube diameters from 850 to 330 μm. The ranges of the gas and liquid superficial velocity were 2.55–322.08 and 0.98–9.78 m/s in the smaller tube respectively. In single-phase flow experiments, the contraction loss coefficients were larger than the experimental results from conventional tubes in the laminar flow. While in the turbulent flow, the contraction loss coefficients were slightly smaller than those from conventional tubes and predicted well by Kc=0.5×(1−σ2)0.75. In two-phase flow experiments, the slip flow model with a velocity slip ratio S=(ρL/ρG)1/3 showed a good prediction that reveals the occurrence of velocity slip. An empirical correlation for two-phase flow pressure drops caused by the sudden contraction was developed based on the proposed contraction loss coefficients correlation for single-phase flow and Martinelli factor.

In several industries, specifically in the petroleum industry, single and multiphase flows through restrictions are commonly found. The restrictions are installed for multiple purposes, e.g., to control the flow and measurement. This research focuses on an experimental study of two-phase (air-water, air-oil) flows through orifice plates, with different restriction sizes (β) to better understand the effects of viscosity in three flow patterns: elongated bubble, slug flow, and annular flow. The experimental setup allows us to visualize the flow with a high-speed camera and measure the differential pressure signature, allowing a detailed discussion on the influence of Reynolds, Froude, and aspect ratio (β) for each flow pattern. This research demonstrates that with simple data analysis (PDF, PSD) of a low-frequency pressure drop sensor installed in the of singularity, it is possible to determine not only the flow pattern upstream the restriction, which has been done before in the literature, but also to determine several properties of the flow upstream and downstream, such as the intermittent factor of the flow and the oscillation wave downstream of the restriction, among others. This paper shows that any restriction properly instrumented can be used to determine the flow pattern in a system, which is usually an unknown variable. The analysis shows how the transition of flow patterns can be observed based on both probability density functions and the power spectrum density of the differential pressure signature.

Predication of wall shear stress of vertical upward co-current adiabatic air-water annular flow in pipes

Gas—liquid annular flow under microgravity conditions: a temporal linear stability study