[en] Highlights: • The effect of inclination on distribution parameter and drift velocity is studied. • Slug behaviors in different flow regions are investigated. • Inclination influences the void fraction and slip ratio. • Drift flux models for calculating void fraction are evaluated. • Correlations of frictional pressure drop are evaluated in different flow regions. - Abstract: A visualization-based investigation was carried out on slug behavior and pressure drop of air–water slug flow in a narrow rectangular duct with cross section of 43 mm × 3.25 mm under inclined conditions. The velocity and length of slugs were obtained through image processing. Based on the liquid Reynolds number, slug flow was divided into laminar flow region (Re_l < 3000) and turbulent flow region (Re_l ⩾ 3000). Experimental results showed that both the slug velocity and the slug length increased with the inclination angle increasing in laminar flow region, while they were nearly unvaried in turbulent flow region. The slug length and slug frequency decreased with the gas superficial velocity increasing or the liquid superficial velocity decreasing for all cases. For laminar flow, the predictions of frictional pressure drop by Chisholm model, Mishima–Hibiki correlation and Lee–Lee correlation could be significantly improved by replacing the void fraction with the ratio of the slug velocity divided by the gas superficial velocity, u_s/j_g; for turbulent region, all three models could well predict the experimental data if void fraction is calculated by the Jones–Zuber correlation

[en] The motion characteristics of the spray droplet as the engineering background were investigated in this paper. The momentum equation of a single droplet in the air environment under room temperature and pressure was established to analyze the sedimentation characteristic, pursuit characteristic and motion trail. The calculation results show that under different initial conditions, there are two situations in which the gravity is greater than the drag force and the gravity is less than the drag force during the initial period of time. After the droplets leave the spray header, the vertical displacement increases rapidly. The larger the droplet size and the initial velocity are, the higher the vertical displacement growth rate is. The larger the difference between droplet size and initial velocity is, the shorter the chasing time and the smaller the chasing displacement will be. The closer the droplet size and initial velocity are, the longer the chasing time and the larger the chasing displacement will be. The larger droplet initial velocity, initial diameter and injection angle induce the slower the transverse velocity disappeared, the larger the lateral displacement reaches, and the larger the area covers by the spray droplet. The calculation results are helpful to optimize the design and layout of spray system in engineering practice. (authors)

[en] Highlights: • Axial additional accelerations are studied under rolling condition. • Factors affecting axial additional accelerations are investigated. • Axial resultant acceleration in rolling motion is studied. • Effect of rolling motion on bubble rising velocity is investigated. • Axial resultant acceleration affects the fluctuation of bubble rising velocity. - Abstract: As a typical ocean condition, rolling motion has a significant effect on the two-phase flow systems. The influences of rolling motion on the bubble rising velocity were studied theoretically and experimentally. Five following rolling conditions were investigated: θ_m5°T8s, θ_m10°T8s, θ_m15°T8s, θ_m15°T12s, and θ_m15°T16s, where θ_m is the maximum rolling angle and T represents the rolling period. The theoretical results show that the fluctuation amplitude of axial tangential acceleration is much larger than that of axial centrifugal acceleration. Although the amplitude of (a_roll)_a is much lower than the value of axial gravitational acceleration, it shows evident influence on the phase curve of axial resultant acceleration. The difference between adjacent troughs in the phase curve of a_a increases as a result of a decreased rolling period or increased rolling amplitude and parameter y′, where y′ represents the horizontal coordinate that is attached to the frame and is in the plane in which the rolling side-to-side motion takes place. The experiments were conducted in a 3 × 3 rod bundle under stagnant condition. The experimental results show that the fluctuation amplitude of bubble rising velocity increases with the increase in rolling amplitude or the decrease in rolling period. In addition, the effect of rolling amplitude is much more significant than that of rolling period. The fluctuation of bubble rising velocity presents one peak in one rolling period under the conditions of θ_m5°T8s, θ_m10°T8s and θ_m15°T8s. However, it presents two peaks under the conditions of θ_m15°T12s and θ_m15°T16s. The axial resultant acceleration gives rise to the fluctuation of bubble rising velocity

[en] The mathematical model describing the influence of periodical inertial forces introduced by rolling on bubbles was presented. The effect of radial forces including inertial and buoyancy forces on bubbles was investigated. Of all the periodical inertial forces, the centrifugal and Coriolis forces can be ignored, while the tangential force plays an important role on bubbles, reaching to its maximum when the rolling platform is at maximal rolling angle. The fluctuation amplitude of periodical inertial forces on radial direction increases with the rolling amplitude and the bubble position away from the channel inlet, and also increases with the decrease of the rolling period. By accounting for the periodical variation of buoyancy force under rolling condition, its amplitude is much larger than those of the periodical inertial forces. The fluctuation amplitude of force increases with the bubble size and is independent with the distortion of the bubble. (authors)

[en] The heat transfer characteristics of spray droplets under medium and low pressure saturated steam environment as the engineering background were investigated in this paper. The results show that the change trends of heat transfer per unit mass with time are basically the same under different initial droplet velocities. The higher initial velocity gives rise to the higher heat transfer coefficient and heat transfer per unit mass. The droplet diameter has a significant influence on the change trend of heat transfer per unit mass with time. The decreasing trend of heat transfer per unit mass decreases gradually with time, and the longer time will be taken for the droplet surface temperature and average temperature reaching to ambient temperature of saturated steam if the droplet diameter is larger. The higher saturated steam pressure gives rise to the more rapid increases of droplet surface temperature and average temperature. Under different steam pressures, the droplet surface temperature or average temperature reach to the saturation temperature nearly at the same time. The surface temperature, average temperature and heat transfer per unit mass change with time basically the same at different initial temperatures. The results are helpful to optimize the design of spray system in practice engineering. (authors)

[en] The experimental investigation of air-water two-phase flow resistance characteristics in a vertical channel with a 3 × 3 rod bundle was carried out under atmospheric and room temperature conditions. Eight classical correlations for predicting frictional pressure drop of two-phase flow were evaluated against the experimental data. The experimental results show that the homogeneous model can predict the experimental data well at high flow rates, but with relatively large deviations at low flow rates. Both the Friedel model and the Lombodi-Pedrocchi model are not suitable any longer for the present case. The Chisholm C model, the Zhang-Mishima model, the Chisholm B model, the Mishima-Hibiki model and the L. Sun model can well predict the experimental data with mean relative errors in the range of 20%-30%. The C factor in the Chisholm C model was modified for giving a new correlation to predict the frictional pressure drop of two-phase flow through rod bundles, showing a good agreement with the experimental data. (authors)

[en] Gas-liquid slug flow is widely encountered in many practical industrial applications. A detailed understanding of the hydrodynamics of the liquid film surrounding gas slug has important significance for modeling of the slug flow. Non- intrusive flow visualization using a high speed video camera system was applied to study liquid film characteristics of slug flow in a vertical narrow rectangular channel (3.25 mm × 40 mm). The liquid film thicknesses on both left and right narrow sides surrounding gas slug are different and wavy, but it has a weak effect on liquid film velocity and the velocities on two sides are equal. The gas and liquid velocity as well as the length of gas slug have a significant effect on the liquid film breaking away thickness. The liquid film breaking away velocity increases with liquid superficial velocity, and decreases with the increase of gas superficial velocity at low liquid velocity. The film does not descend and the gas superficial velocity has no significant effect on liquid film breaking away velocity at high liquid velocity (j_L≥1.204 m/s), it is mainly determined by the liquid flow rate. (authors)

[en] A visualized study of flow patterns was carried out for air-water two-phase flow in two narrow rectangular ducts (40 mm × 1.41 mm, 40 mm × 10 mm) respectively. The superficial velocity ranges of air and water were 0.01-0.59 m/s and 0.02-3.72 m/s. Based on three classical criteria on transition from bubbly flow to slug flow, critical transition void fraction 0.23 was deduced by introducing the effect of channel size. The rectangular ducts were classified as small channel (s≤2.5 mm) and conventional channel (s>2.5 mm), based on which a modified transition criterion for bubbly flow to slug flow was proposed. The modified criterion shows a good agreement with the experimental data from this study. Experimental data from Mishima, Wilmarth and Sadatomi etc. were also collected to further evaluate the accuracy and applicability of the modified criterion, showing a good prediction to these experimental data as well. (authors)

[en] Visualized investigation of the air-water two-phase slug flow in vertical narrow rectangular channel (3.25 mm × 40 mm) was carried out with the help of a high speed video camera. The work pressure is atmospheric pressure and the superficial velocities of air and water cover 0.1-2.51 m/s and 0.16-2.62 m/s, respectively. The slug flow in narrow rectangular channel is found to be much different from that in circular pipe. Most of the slug bubbles are deformed, and much more seriously at high liquid velocity. The void fraction of narrow side liquid film is larger than that of wide side liquid film, and the film does not descend at high liquid velocity. It is observed that some tiny bubbles in the areas of the nose and tail of a slug bubble are injected into the liquid film along the wide side. The nose shape of the slug bubble as well as its width has significant effect on its velocity, whereas the effect of the length is very weak. The Ishii and Jones-Zuber drift flux model can predict slug velocity well, except at low liquid superficial velocity by the reason of that the calculated drift velocity is less than experimental values. (authors)

[en] Using air and deionized water as working fluids, flow pattern characteristics of two-phase flow was investigated experimentally for two vertical rectangular channels with the cross sections of 1.41 × 40 mm"2 and 10 × 40 mm"2. Experiment was conducted under atmospheric condition, and the superficial velocities of air and water were in the ranges of 0.015 ∼ 0.59 m/s and 0.025 ∼ 3.74 m/s. Four classical transition criteria from bubbly flow to slug flow are evaluated against present experimental data and that from open literatures. The result shows that the four transition criteria are limited to some extent. Through the analysis of the dataset, it is found that aspect ratio is a key parameter to determine the critical transition void fractions for which two correlations are given according to the threshold of 10 mm of the hydraulic diameter. As a result, the modified criterion is obtained and is satisfactory agreement with the present data and the existing experimental data, and it is better than four classical transition criteria. (authors)