Churn Flow Research Articles

Study on the Flow Pattern and Transition Criterion of Gas-Liquid Two-Phase Flow in the Annular of Shale Gas Fractured Horizontal Wells

Improving the accuracy of pressure prediction in the wellbore annular is of great importance for the design in oil and gas production. However, due to the existence of double-layer liquid membrane and the lack of relevant experiments, the existing correlations fail to the field application. In this study, a new model of flow pattern transition in inclined annulus pipe is proposed by using a mechanistic approach to classify the flow patterns. Firstly, a gas-liquid two-phase flow experiment in annulus pipe was carried out in a pipe with an outer diameter of 73.02 mm and an inner diameter of 121.36 mm, and then the influence of inclined angle on the transition boundary of flow pattern is discussed. Finally, a hydrodynamic transition criterion for the flow pattern model of inclined annulus pipe is established and verified in detail. The experimental results show that bubble flow, slug flow, churn flow and annular flow were observed under different inclination angles, and the results indicate that the slug flow will be shifted to the larger gas-liquid superficial flow rate region with the smaller inclination angle, and the annular flow will appear in the higher gas superficial flow rate region. Compared to the performance of the existing correlations (Kelessidis and Zhang) and the present model using the experimental data, the accuracy of the new model reached 83%, significantly higher than the other two models, and the new correlation was better in predicting the transition from slug flow to churn flow and churn flow to annular flow.

Two-phase flow pattern identification in CAES systems with dimensional analysis coupled with support vector machine

Two-phase bubbly flow can lead to challenges during the charging period for hydraulically-compensated Compressed Air Energy Storage systems. In this study, upward two-phase flow systems have been investigated in large diameter vertical pipes to better understand when the transitions to problematic flow regimes occur. A comprehensive dimensional analysis was conducted, considering vertical pipe geometry and fluid thermophysical properties. Nine dimensionless parameters and three two-phase flow maps were created. The plot of liquid Reynolds number, ReL, over the Two-phase Reynolds number, Retp,made classification between the Bubbly, Slug, and Churn flow difficult for large diameter pipes. The plot of the two-phase Capillary number,Catp, over the liquid Reynolds number, ReL, showed large overlap of the data sets extracted from the literature. This may have been due to the weakness of the capillary number in defining the larger pipe flow behaviours. The last flow map based on the Slippage number, Sltp, versus the two-phase Froude number, Frtp, showed an exponential decay relationship between the Froude and Slippage parameter. When this was converted to the semi-log scale it provided a valuable classification map of two-phase flow. Finally, a Support Vector Machine (SVM) technique applied to the 2D plot of (Sltp- Frtp), gave a useful transition boundary to separate regimes in air-water upward flows in large diameter pipes.

Experimental investigation on two-phase flow noise characteristics of R410A through electronic expansion valve of multi-split air conditioner

The refrigerant flowing through an electronic expansion valve (EEV) of a multi-split air conditioner is usually in two-phase state which may generate more severe flow noise than that in sub-cooled state. The purpose of this paper is to experimentally investigate the influences of the two-phase R410A operation conditions on the flow noise characteristics in the EEV, including listening sense forms, frequency distributions and sound pressure levels (SPLs). The experimental conditions cover the refrigerant mass flow rate ranging from 30 to 90 kg h–1 and the refrigerant quality ranging from 0.02 to 0.4. The results show that low mass flow rate facilitates the reduction of noise frequency, and the peak noise frequency was reduced by up to 32% as the mass flow rate decreased. Low refrigerant quality helps to reduce the SPL of flow noise, and the average SPL decreased by 6 dB as the refrigerant quality decreased. The guidance of inlet flow pattern by transforming the churn flow to the annular flow helps to simultaneously reduce the noise frequency and SPL, and the peak noise frequency as well as the averaged SPL decreased by 47% and 4 dB, respectively. Moreover, the hissing noise and the gurgling noise are the only two kinds of listening sense forms of flow noise detected in the present experiment, and the peak frequency and the average SPL of hissing noise are both higher than those of gurgling noise.

Experimental study of flow boiling characteristics in minigap channels over a wide heat flux range

The flow boiling characteristics in minigap channels were investigated by observing the two-phase flow patterns, measuring the two-phase pressure drops, and calculating the two-phase heat transfer coefficients and critical heat fluxes in minigap channels with heights of 0.5–2 mm. The working fluid was deionized water with mass fluxes of 200–400 kg m−2 s−1. The results show that bubbly flow, sweeping flow, churn flow, and churn-annular flow occur in sequence with increasing heat fluxes. An unstable two-phase flow regime was observed during the sweeping flow. The pressure drop varied in three regimes that were related to the flow patterns. The maximum pressure drops are 12.2, 3.8 and 1.6 kPa, respectively in 0.5 mm, 1 mm and 2mm channels at 400 kg m−2 s−1. The effect of gap height on the heat transfer coefficient first increases and then decreases with increasing heat flux. The heat transfer coefficients of all channels are at a high level when the heat flux is high. The maximum heat transfer coefficients are 7.3, 6.9 and 5.8 W cm−2 K−1 in the 0.5 mm, 1mm and 2mm channels at 400 kg m−2 s−1, which occurred before critical condition. The critical heat flux increased with increasing mass flux and gap height. In addition, a pressure drop correlation was developed to more accurately predict the pressure drop in microgap channels with an accuracy of 10.8%. The Liu-Winterton correlation and the Shah correlation give the best predictions for the subcooling boiling heat transfer coefficient. The Sun-Mishima correlation and the Kandlikar correlation provide the most accurate predictions for the saturated boiling heat transfer coefficients.

Evolution of Gas-Liquid Two-Phase Flow in an M-Shaped Jumper and the Resultant Flow-Induced Vibration Response

The vibration excited by gas-liquid multiphase flow endangers the structural instability and fatigue life of subsea jumpers due to the cyclic behavior. In this paper, the multiphase flow-induced vibration (MFIV) of an M-shaped jumper is numerically investigated using a two-way fluid-structure interaction (FSI) approach. The effect of gas-liquid ratios (β) ranging from 1:1 to 1:5 is examined with a fixed flow velocity of 3 m/s, and the influence of mixture velocity (vm) in the range 2–6 m/s is evaluated with a gas-liquid ratio of 1:1. The numerical results reveal the detailed flow evolution of the gas-liquid mixture along the jumper. With inflow of slugs, the pattern successively experiences the slug flow, wavy flow, imperfect annular flow, stratified flow, churn flow, wavy flow and imperfect annular flow in the pipe segments when β = 1:1 and vm = 3 m/s. This development of mixture flow is significantly altered by changing either the gas-liquid ratio or the mixture velocity. In comparison with the flow evolution in a stationary jumper, the pattern in each pipe segment is not been substantially changed due to the limited response amplitude of order of 10−3D (D is the outer diameter of the jumper). Due to the complex flow evolution, the pressure acting on the six bends of the jumper fluctuate in multiple frequencies. Nevertheless, the dominant fluctuation frequency is approximately equal to the inflow slug frequency. Moreover, the inflow slug frequency also dominates the in-plane response of the jumper. Both the in-plane and out-of-plane responses of the jumper exhibit spatial-temporal variation characteristics. The most vigorous oscillation occurs at the midspan of the jumper. As β is reduced, the out-of-plane response of the jumper midspan is suppressed while the in-plane response is enhanced. In contrast, both the in-plane and out-of-plane oscillations of the jumper midspan are amplified with the increase of vm.

Gas-liquid flow regime identification via a non-intrusive optical sensor combined with polynomial regression and linear discriminant analysis

The identification of gas-liquid flow regimes for co-current upward flow in a vertical pipe is investigated using Polynomial regression (PR) and Linear Discriminant analysis (LDA) on the response from a non-intrusive optical sensor. The average voltage and associated standard deviation can be extracted as features from the sensor and used as inputs to establish the coefficients of an optimal regression model. Further to this, due to the structural similarity between the slug and churn flow regimes, further discrimination was required in the form of a Linear discriminant analysis to give accurate classification between these two flow regimes. According to the offline test and real-time flow conditions considered, a classification accuracy of 100% is achieved. The results of this paper show the capability of applying a relatively simple supervised regression model and LDA to the response from a non-intrusive optical sensor to identify gas-liquid flow regimes efficiently and accurately.

Solid particle erosion prediction in elbows based on machine learning and swarm intelligence algorithm

Continuous impact of solid particles causes severe pipeline wear, and may result in leakage in directional change areas such as elbows. Accurate prediction of erosion is essential in the petroleum industry. In this study, a new machine learning (ML)-based model for predicting elbow erosion was established. Seven parameters were selected as feature inputs from the fluid characteristics, particle characteristics, and pipe characteristics. The maximum erosion rate was used as the predicted output. Based on the gas–solid flow data, the prediction accuracy of different ML models was compared. The kernel extreme learning machine (KELM) model was considered as the optimal model. For gas–liquid–solid flow, incorporating swarm intelligence (SI) algorithms, the whale optimization algorithm–hybrid kernel extreme learning machine (WOA–HKELM) erosion prediction model was proposed; the predictions were compared with the experimental values. The root mean square error (RMSE) of the prediction was 0.82 × 10−3, which is consistent with the experimental results. It was also demonstrated that the model can capture the trend of the influence of mixed dimensionless inputs on the prediction results in churn and annular flow.

Numerical Simulation of the Flow Pattern of Spiral Annular Flow with a Guide Strip by Spiral On-Way.

To study the gas–liquid two-phase spiral annular flow pattern and its conversion law in a horizontal tube with twist tape by spiral on-way, a numerical simulation is carried out using the RNG k-ε model, and the DPM model is adopted as the particle motion model. The research results show that three main flow patterns are obtained, which are spiraling churn flow, spiraling annular flow, and spiraling annular twine flow. The ratio of the liquid-phase to gas-phase velocity VL/VG is the main factor that affects the existence of the gas phase. When VL/VG > 1/2, the flow pattern is a spiral cluster flow; when 1/5 < VL/VG< 1/2, the flow pattern is a spiral annular flow; and when VL/VG < 1/5, the flow pattern is a spiral annular flow. Compared with the flow pattern of the local spiral flow pattern of the short twisted band, it was found that the occurrence conditions of annular flow were postponed and a new flow pattern appeared. The research results of this paper have guiding significance for the pipeline risk management of natural gas hydrates and the engineering application of the full rotation of guide strips.

Experimental study of two-phase flow patterns through an inclined rectangular channel with horizontal and vertically upward directions

In this paper, an experiment is carried out to recognize the flow pattern and to measure the void fraction of gas–liquid two-phase flow in an upward inclined channel (horizontally to vertically upward). The experiments are performed in a 30-m channel with dimensions of 100 × 150-mm utilizing air and water as a fluid mixture. Three flow patterns, including slug, churn, or annular-mist flow are observed in the vertical section. It was determined that upstream impacts influence the interactions as well as momentum transfer among the two phases. Also, the amount of backward flow consequence of the formation of counter-current flow across the vertical section is specified by downstream impacts. The outstanding regime of flow in the vertical section was from the churn flow regime. The flow regime through the first horizontal section is classified into two parts: the fully developed part and the inlet bend section which is affected by disturbance pursued by the mixer and vertical section.

Flow Instability of a Liquid Through a Small-Scale Channel During a Flow Evaporation

Finite volume analysis of two-dimensional model has been performed to study of the influence of heat flux and mass flux on two phase heat transfer through microchannel during water flow boiling. The channel dimensions were (5×5××50) m3. The inlet boundary conditions were (1.63, 3.32, 4.9, 6.65, 8.16) . The supplied heat fluxes were 24.8 , 33.2 , 46.13 , 51 and 69 for nucleate boiling while 33.3 , 50 , 66.7 , 83.3 , and 100 for dryout condition. The inlet temperature was 30oC. It was shown that the highest heat transfer coefficient occurs at beginning of onset of nucleate boiling (ONB) after that the it will drop due the changing of the flow patterns. According to supplied heat flux, the patterns were bubble flow, dispersed bubbles, churn flow, annular flow, and then dry out condition occurs. For the mass flux 1.63kg/ the highest heat transfer coefficient was 9.5 kW/ while for the mass flux 8.16 kg/ the highest heat transfer coefficient was 20kW/ . For the mass flux 1.63kg/ the highest value of vapour quality was 0.09 while for the mass flux 8.16 kg/ the highest value was 0.15. There is a sharp pressure drop when starts to nucleate and the highest value occurs at dryout condition. This study gave an understanding of the heat exchange flow behaviour during water boiling through a single microchannel.

Vibrations of a vertical 90° pipe bend under two-phase flow conditions

PurposeA simulation framework that includes a finite element analysis (FEA) and computational fluid dynamics (CFD) model is generated to study the effect of unstable two-phase flow-induced vibrations at a vertical 90° pipe bend. The corresponding fluid-structure interaction (FSI) of an unstable flow may pose danger to the piping structure. This paper intends to discuss this interaction.Design/methodology/approachFour cases of flows under the slug flow and churn flow regimes were investigated. The flow regimes vary in superficial gas velocities with velocities from 0.978 m/s to 9.04 m/s, while the superficial liquid velocity is kept constant at 0.61 m/s. The pipe model consists of an internal diameter of 0.0525 m, a bend radius of 0.0762 m, and a stainless-steel pipe structure.FindingsResults show that the average unstable void fractions increase with the superficial gas velocities, but the peak frequencies were constant at 13 Hz for three of the cases. The total displacement and von Mises stress increase with a declining rate in each subsequent case, while the RMS of von Mises stress begins to stall at superficial gas velocities between 5 m/s and 9.04 m/s. The peak frequencies of von Mises stress decrease in each subsequent case.Originality/valueThe proposed model can be used to investigate the FSI effect of unstable void fractions at pipe bends and could assist in the development of piping systems in which the use of piping elements arranged close together are unavoidable.

The effect of decay on churn flow transition in a vertical gas–liquid swirling flow

Gas–liquid two-phase swirl flow has been widely used in industry, especially the decayed swirl flow. However, axial development of the decayed swirl flow which is closely related to the effective length of swirl are lack of understanding. In this work, effects of decay on churn flow transition induced by a vane-typed swirler were investigated by a visualization experiment in a long vertical pipe (inner diameter of 62 mm and total length of 11 m). The results show that under the effect of swirl, churn flow can transform to a swirling annular flow, meanwhile the pressure drop increases and its fluctuation weakens compared with these in the non-swirl churn flow. However, the length of the transformed swirling annular flow is limited, due to the decay of swirl intensity. Then the effect of swirl decay on churn flow transition has been analyzed by the theoretical model proposed by our previous work (Liu et al., 2019). Finally, a new method was proposed here to predict the effective length of swirling annular flow transformed from churn flow in the vertical pipe, by combining the swirl decayed model and churn flow transition model in the swirl flow.

Effect of Oil Viscosity on Flow Pattern Transition of Upward Gas-Oil Two-Phase Flow in Vertical Concentric Annulus

Summary Gas-oil two-phase flow usually occurs during gas influx into the wellbore annulus of deepwater drilling with oil-based drilling fluid or gas-oil production from the wellbore annulus. A flow pattern transition often is used in well control or in production parameter design. Related studies primarily have focused on gas-water two-phase flow, which is significantly different from gas-viscosity oil two-phase flow. We conducted experiments on a facility that included a flow section that was 13 000 mm long, with a 60-mm inner pipe and a 100-mm outer pipe in the annulus. A range of oil viscosities from 16 to 39 mPa∙s has been studied. The superficial gas and oil velocities varied from 0.55 to 17.077 m/s and from 0 to 0.414 m/s, respectively. A flow pattern was identified based on visualized analysis using electrical capacitance volume tomography (ECVT) and void faction wave analysis. Flow pattern maps of different viscosities were plotted based on measured data. Transition criteria based on Froude number and Archimedes number were established. The new model was compared with gas-water two-phase flow transition models and validated with experimental data. The effect of viscosity on flow pattern transition was revealed. The transition boundary of the bubble to slug flow and slug to churn flow both moved in the direction of smaller superficial gas velocity with an increase in oil viscosity. There may exist a critical viscosity value when churn flow transited to annular flow. When the viscosity was lower than this critical point, the above result was the same for churn to annular flow when the superficial oil velocity was low. With an increase in the superficial oil velocity, however, the boundary gradually changed to high superficial gas velocity as the oil viscosity increased. When the viscosity was larger than the critical point, the oil viscosity had a slight influence on the transition.

A new deep neural network framework with multivariate time series for two-phase flow pattern identification

Uncovering flow dynamic behavior of different flow patterns is an important foundation of multiphase flow research. But the traditional classifier is still adopted in the flow pattern identification based on statistical features of experimental measurements, and the utility of data is not sufficient in previous works. Therefore, a novel deep neural network framework is proposed to leverage abundant details of signals. The data was input into the new model after two innovative slicing operations, which combines BiLSTM and CNN to extract the deep characteristic information of different flow patterns. In addition, attention mechanism and residual connection are introduced to improve the network performance. Meanwhile, the dynamic experiment of vertical gas-water two-phase flow is carried out, four-channel conductance signals under five typical flow patterns, namely bubble flow (BF), slug flow (SF), bubble-slug transitional flow (BSF), churn flow (CF) and slug-churn transitional flow (SCF), are collected to feed the network. Finally, in order to verify the effectiveness of the proposed model, some comparative experiments are designed and implemented. The results demonstrate that our proposed model outputs more precise flow pattern identification, which opens up a new way for investigating industrial multiphase flow.

Significant Differences in the Value of the Void Fraction in Various Types of Low-Viscosity Two-Phase Flow Patterns in Capillary Tubes with a Particular Slope

A two-phase flow is the simplest form of multi-phase flow. Two-phase flow has basic parameters, including flow behavior, flow pattern, void fraction, pressure gradient, or pressure drop. The development and application of fluid flow in small channels (mini and micro) are in the Micro Electromechanical System (MEMS). In general, the application of two-phase flow can be found in various applications in the industrial world, such as boilers, cores, and steam generators in nuclear reactors, petroleum transportation, electronic cooling, and different other types of chemical reactors. One of the most common challenges is void fractions and flow patterns in horizontal and vertical pipelines. The purpose of this study was to find out the significant difference in the characteristic value of fundamental parameters, primarily void fractions and flow patterns that occur in capillary pipes with a slope of 15o to horizontal. The fraction value and shape of this flow pattern are essential because it determines the pressure drop value, which means determining the level of danger that can occur if the pressure drop value reaches the extreme value. The fluid used a mixture of air-water and glycerin with 0, 10, 20, and 30% concentrations. To determine the void fraction is used digital image processing method with MATLAB R2014a application. A 1200 fps high-speed camera and 640 x 480-pixel resolution are used for video shooting. The experiment was conducted at superficial gas (JG) velocities of 0.025 - 66.3 m/s and superficial fluid velocities (JG) intervals of 0.033 – 4.935 m/s. The experiment results showed a significant difference between the void fraction value for the bubbly flow pattern and the churn flow pattern, which has not been widely discussed in previous studies. The analysis concluded that the higher the velocity of superficial gases, the void fraction value increases. This increase in the void fraction is why there is a significant change in flow patterns from bubble flow patterns to churn. In addition, it was also found that the increased viscosity of the fluid significantly affected changes in flow patterns, incredibly bubbly and plug, but its velocity decreased. It is also the result that the higher homogeneous void fraction (β) value affects the increase in the length of bubbly flow patterns and plugs.

Liquid Loading in Natural Gas Vertical Wells: A Review and Experimental Study

Summary Liquid loading in natural gas wells is one of the main causes of decline and eventual cease of production. Identifying the onset of liquid loading and its impacts on production is vital in production optimization of gas wells. In this study, a new state-of-the-art experimental facility is used to analyze two-phase flow parameters, such as liquid and gas velocities, pressure drop, liquid holdup, and flow pattern, and study the onset of liquid loading. Experiments are conducted using air and water in a 0.0508-m (2-in.) internal diameter (ID) vertical flow loop. The facility is designed to eliminate any source of disruption to the flow in a temperature-controlled indoor environment to avoid changes in fluid properties. A wide range of data are collected related to the onset of liquid loading and in the churn region. The acquired data are compared with other studies to ensure the accuracy and repeatability of the tests. A quantitative method is developed to predict the onset of liquid loading, namely, the positive frictional pressure gradient (negative pressure drop). A comparison is made between the visual observations for flow pattern transition, positive frictional pressure gradient, and minimum pressure drop. Overall, the results show that the positive frictional pressure gradient approach can provide a better estimation of the onset of liquid loading than the minimum pressure drop approach. Additionally, Tulsa University Fluid Flow Project (TUFFP) unified model (v2016) and OLGA (v2016.2.1) are used to evaluate the average pressure drop, liquid holdup, and flow pattern. Furthermore, the liquid film reversal model, liquid droplet models, and inflection point approach are included in the comparison of the onset of liquid loading with experimental data. The OLGA model predicts the results more precisely than other models. The comparison with the experimental data indicates that inaccurate flow pattern predictions could significantly increase the relative errors in pressure drop and liquid holdup. In addition, the evaluation of OLGA results suggests the need to develop a model for churn flow, as the relative error increases sharply in the churn flow with the model predicting the transition to slug flow.

Image identification for two-phase flow patterns based on CNN algorithms

Flow patterns are essential and useful to model the interfacial structures and heat transfer in gas-liquid two-phase flow. However, the current two-phase flow patterns classification methods mostly depend on direct visual observation. This study adopted a new flow pattern classification method based on convolutional neural network (CNN) algorithms to achieve an automatic and objective identification of two-phase flow patterns. A database of 696 test conditions, including 105642 condensing flow pattern images of methane and tetrafluoromethane in a horizontal circular tube, is collected as the input of the data-driven algorithms. After 80% of image data is fed to train and fit the parameters in the algorithms, the trained models with acceptable universality are obtained to identify five flow patterns: annular flow, bubbly flow, churn flow, slug flow and stratified flow. Compared with the manual classification, the proposed method can accurately predict two-phase flow patterns with a prediction accuracy of more than 90.63% and 91.45% for the test dataset and the entire database, respectively. The average accuracy for predicting all data points in the database is more than 97.56%. The results showed that using images as input, CNN algorithms can provide objective prediction with satisfactory accuracy and universality for two-phase flow pattern identification.

Visualized-experimental investigation on a mini-diameter loop thermosyphon with a wide range of filling ratios

The mini-diameter loop thermosyphon, which has an excellent heat transfer performance with simple, compact structure and pump-free advantages, can effectively solve the problem of heat transfer under high heat load in a confined space. The mini-diameter loop thermosyphon exhibits different gas-liquid transport and heat transfer patterns than the conventional-diameter one due to various gas-liquid phase transformation properties and a strong friction force in the mini-diameter channel. Furthermore, data on the visual analysis of gas-liquid transport and heat transfer properties of the mini-diameter loop thermosyphon under various filling ratios is still lacking. In this paper, a visible experimental research on a mini-diameter loop thermosyphon (3 mm) in a wide filling ratio range(44% -97%) is undertaken. The maximal heat transfer capacity increases initially and subsequently falls with the filling ratio, according to the testing data. At a low filling ratio, the flow pattern changes as slug flow - churn flow - annular flow with the heat load increase, leading to a low loop thermal resistance, and dry out occurs at 185 W heat load under 50% filling ratio. The gas-liquid flow shifts to slug flow-geyser boiling-bubble flow at high filling ratios (73% and 82%), and the bubble departure diameter steadily decreases with increasing heat load. Consequently, the heat transfer performance (thermal resistance) has been enhanced, and the maximum heat transfer capacity has been increased to 280 W. Single-phase flow is plainly seen in the loop when the filling ratio reaches 91% or even 97%, thus reduced heat transfer capacity. Therefore, the high-filling-ratio mini-diameter loop thermosyphon is better for high-load heat dissipation. However, a low thermal resistance shows at the low filling ratio, but the heat load is also low. As a result, a loop thermosyphon with a low filling ratio should be selected for the low heat load situation.

Identification of two-phase flow regime in the energy industry based on modified convolutional neural network

The identification of the two-phase flow regime is the most fundamental and crucial target of fluid mechanics in the nuclear thermal-hydraulic analysis since the accuracy of fluid and heat transfer models depends on the reasonable prediction of flow regimes. The trend is quantitative identification based on the measurement of quantifiable parameters or the quantitative information of the two-phase flow images. Machine learning is currently the mainstream methodology for flow regime identification. The state-of-the-art method is using flow regime images for identification based on convolutional neural networks (CNNs). To achieve a higher classification accuracy and overcome the disadvantage of the traditional CNN (time-consuming owing to a large number of images required for training), this paper proposes a more efficient methodology for two-phase flow regime identification, which combines the advantages of CNN for image feature information extraction and the support vector machine (SVM) for classification - the CNN-SVM method. Furthermore, the traditional CNN needs a large number of images for training. Considering the limited flow regime dataset, the CNN-SVM method in this paper uses the idea of transfer learning. A small-scale flow regime dataset is used for validating the effectiveness of the proposed method. The accuracies for SVM, CNN and CNN-SVM were 93.9%, 97.2% and 98.8% respectively, which show that the proposed integrated method can realize better classification results since it combines the capability of strong feature extraction of CNN and classification techniques of SVM. The improvement is mainly reflected in the higher classification accuracy of slug flow and churn flow. Additionally, according to the results, the proposed method is also more robust.

Particle Transport Velocity in Vertical Transmission with an Airlift Pump

This paper presents the optimal conditions for fast transfer of solid particle with an airlift pump. The experimental examinations were carried out in an airlift pump with a length of 5.64 m and an inner diameter of 0.102 m in order to determine the impact of submergence ratio, air flow, and physical particle properties, such as shape, size, and density, on the vertical velocity of the particle in detail. The results showed that with the same air flow, the maximum particle velocity was achieved when the churn flow regime is established with a submergence ratio close to 0.89. However, in bubble and slug flow, it is not possible to carry a large particle in the dimensions of centimeters. Furthermore, in a churn flow, the velocity of the particle exceeds the velocity of pumped water; hence, water is not the only particle carrier in a vertical three-phase flow.