Search

• The Enkf algorithm was developed and validated. • Δ p u 7 / 4 has been verified in the narrow rectangular channel. • Multiple iteration improves the accuracy of Enkf more effectively than the ensemble size. • The k-epsilon model assimilated with pressure drop exhibits improved predictive capability for velocity distribution. Due to significant differences in channel geometry, standard turbulent models cannot accurately predict the flow characteristics of narrow rectangular channels (NRC). In this paper, a benchmark method of turbulence model using Enkf-based (Ensemble Kalman filtering) data assimilation and scalar experimental data was developed. The evolution of pressure and the velocity profile were experimentally obtained. The algorithm of the Enkf was established and validated, which indicates that multiple iteration improves the accuracy more effectively than simply increasing the ensemble size. Using the developed approach, the initial parameters of the standard k-epsilon model were benchmarked and the accuracy was carefully explored. According to the results, the pressure drop within the NRC is approximately proportional to the 7/4 power of the velocity. The accuracy of the corrected k-epsilon model has been significantly improved. The comparison results of the pressure evolution and velocity profile further illustrate the effectiveness and robustness of the developed method. [ABSTRACT FROM AUTHOR]

[Display omitted] • Two kinds of two-phase stereo-PIV measurement techniques were developed. • The image processing and velocity reconstruction methods were introduced. • The simultaneous measurement of 3D velocity and bubbles was achieved. • The accuracy of two-phase stereo-PIV was valid by both 2D-PIV and DNS data. The bubbles and three-component (3C) velocity are of great significance in the investigation of gas–liquid two-phase flows. In this study, two kinds of two-phase stereo-PIV measurement techniques, based on optical phase discrimination and image phase discrimination, were proposed to realize the simultaneous measurement of 3C velocity and bubbles. The corresponding image processing and velocity reconstruction methods were developed. Three validation tests were conducted to verify the accuracy and feasibility of the proposed techniques. The measurement results of the pipe flow and the motor-driven swirl flow illustrate the sufficient accuracy of the techniques. Finally, the techniques were applied to the vertical bubbly pipe flow and the rising bubble in the bubble column. The results show that both OPD and IPD methods have successfully obtained the three-dimensional velocity and bubbles. Through comparison, the IPD method was greatly recommended for two-dimensional (2D) measurement due to its low equipment requirements, simple setups, and high spatial resolution. [ABSTRACT FROM AUTHOR]

• The characteristics of bubbles and turbulence in the NRC have been studied. • The bubble size effect has been studied independently by using the CSBG. • u y ′ / u x ′ are 1.9, 1.5 and 2 in single-phase, large bubbles and small bubbles cases. • The modulation of turbulent intensity by the bubbles is characterised well by ϕ = 1. Narrow rectangular channels are widely used in high heat flux applications, due to the high heat transfer efficiency and compact structure. However, a rare understanding was known about the two-phase turbulent characteristics. In this study, an experimental investigation of bubble and turbulent characteristics of dispersed bubbly flow in the narrow rectangular channel was carried out by using the high-speed camera and two-phase particle image velocimetry. The bubble size effect was studied independently using the continuous spectrum bubble generator. According to the results, the bubble size distribution conforms to a Weibull distribution, and the wall-peak void fraction distributions were found. Both turbulent enhancement and suppression were observed in current conditions. Large bubbles enhance turbulence intensity, while small bubbles suppress it. The eddy viscosity ratio ϕ characterizes the modification of turbulence intensity by bubbles, with a critical value of ϕ = 1. [ABSTRACT FROM AUTHOR]

• An identification method of overlapping bubbles in high void fraction conditions by Convolutional Neural Network was proposed. • A trajectory tracking technology for overlapping bubbles was achieved based on Kalman Filter. • The model was trained only by synthetic images generated by GAN. • Overlapping bubbles can be detected and tracked accurately under low illumination and strong noise conditions. Gas-liquid bubbly flow is widely applied in chemical process engineering. Geometric and dynamic parameters of bubbles play an essential role in the numerical prediction of mass and heat transfer processes. However, the critical obstacle in bubble detection is the inability of bubble segmentation and reconstruction when the overlapping issue of multiple bubbles is serious under high void fraction conditions. A new detection and tracking technique for overlapping bubbles was proposed in this paper to identify the overlapped bubbles. First, a novel convolutional neural network is used to detect bubbles. Afterward, the relationship between the detected bubbles in two frames is correlated using the Kalman Filter and neural network. The algorithm achieves 85 % accuracy under high overlap rate conditions in a 10 mm narrow rectangular channel with around 0.1 s for an image. In addition, a comparison test was conducted to evaluate the present technique's accuracy and robustness compared with conventional methods. [ABSTRACT FROM AUTHOR]

• The CSBG, which can produce bubbles with continuous sizes is developed. • Experimental results show that CSBG can generate bubbles of continuous sizes. • Overlapping bubbles processing technique (OBPT) based on CPD & FSAC is proposed. • Accuracy of OBPT exceeds 95% with a sub-second response time per hundred bubbles. The parameter of bubble size is of great significance to the bubbly flow. In this paper, the Continuous Spectrum Bubble Generator (CSBG), consisting of a sparger and impeller, was developed to realize the continuous control of bubble sizes. By introducing a new variable, rotation rate of the impeller, bubble size and gas superficial velocity are successfully decoupled. Experimental results show that bubble sizes continuously drops from 3 to 1 mm, without changing gas superficial velocity. Also, the spectrum of bubble size can be expanded by increasing or decreasing mass flow rates. To detect the overlapping bubbles, a new overlapping bubbles processing technique based on fast segmented arcs clustering was proposed. It's efficient to cluster the segmented arcs through detecting the Major and Pair arcs and connecting the Multi-segment arcs through concave points connection methods. Test results show that the accuracies of mean diameter exceed 95% with sub-second response time per hundred bubbles. [ABSTRACT FROM AUTHOR]

Tip cavitation remains a complex and critical concern for waterjet pumps, affecting the safety and efficiency of pump operation. This paper aims to explore the cavitation scale's evolution law, the cavitation structure's evolution characteristics and mechanisms, and how these factors affect pump performance. A synchronous experimental system consisting of a test pump, a high-speed camera, and pressure transducers has been established. Results show that the tip cavitation mainly consists of primary leakage vortex cavitation (PLTVC), secondary leakage vortex cavitation (STLVC), and suction surface attached cavitation near the blade tip (SSAC). The gray level-based image processing method is used to reveal the law of cavitation scale growth rate and vapor fraction distribution. The tip cavitation scale's growth rate increased as the cavitation number decreased. The rapid expansion of the attached cavitation promotes the formation of a re-entrant jet, cavity shedding, and the formation of perpendicular cavitating vortices (PCVs). It was then found that for every 0.1 decrease in cavitation number, the PCV scale in the blade overlap region increased by about 14.1 % during σ =0.256–0.206, which severely blocked the tip passage flow and accelerated the degradation of pump performance. The spatial-temporal evolution of the PCVs is classified into three stages: the shedding stage of the attached cavity, the stage that the shedding cavities change the migrating direction and form the PCVs, and the collage stage of the PCVs. With cavitation numbers of 0.256 and 0.223, the PCV evolves at roughly 194 Hz and 144 Hz, which results in pressure fluctuations of approximate frequency. [ABSTRACT FROM AUTHOR]

• Idea of indirect quantification of interfacial mass transfer velocity k l of bubbles under high Reynolds number conditions. • SI-LIF measurement technique and reconstruction algorithm to attain physical quantities required to derive k l. • 3-D visualization and quantification of mass transfer of single bubbles rising in both quiescent water and homogeneous turbulence. • The measured k l are independent of R e T , and all collapse to the analytical solution of potential flow approximation. • Dominant effect of buoyancy of bubble instead of turbulent eddies on the interfacial mass transfer process is confirmed across the hydrodynamic conditions presented in this paper. Accurate quantification of the mass transfer coefficient k l of bubbles under high Reynolds number conditions has been in desperate need for decades to provide insights into its fundamental physics and reach a general mass transfer model. In this paper, a state-of-the-art SI-LIF technique is developed to quantify k l indirectly based on the law of conservation of mass. The reconstruction algorithms of the required physical quantities are validated numerically, suggesting an overall uncertainty of less than ± 5 % in the estimation of k l. We perform mass transfer experiments of single oxygen bubbles with d eq = 1.5 mm rising in both quiescent water and homogeneous turbulence, with turbulent Reynolds number R e T ranging from 0 to 1114. The measured k l under turbulent conditions are independent of R e T , and all collapse to the analytical solution of potential flow approximation, revealing that the flow in the immediate vicinity of the bubble interface remains undisturbed by the fluctuating liquid motions across all the turbulence settings examined in this paper. [ABSTRACT FROM AUTHOR]

• The TFM-PBM method is implemented to study the hydrodynamics and interfacial mass transfer characteristics of bubbly flow in a novel serpentine tubular reactor. • Simulation results are in reasonable agreement with measured data in terms of the evolution of bubbly flow and overall efficiency of mass transfer. • Different bubble breakup and coalescence model are tested in the curved flow. • Mass transfer characteristics predicted by the slip penetration model and eddy cell model are studied. The multiphase behavior in gas–liquid reactors is a focused topic in chemical engineering. In this paper, the TFM-PBM method is implemented to study hydrodynamics and interfacial mass transfer characteristics of bubbly flow in a novel serpentine tubular reactor. The calculated evolution of bubble diameter and the gas phase distribution which converts between the 'wall-peak' and 'core-peak' pattern while flowing across the serpentine tube are in reasonable agreement with experiment results. Both the slip penetration model and the eddy cell model are employed to investigate the local mass transfer process, and the overall k l a predicted by the two distinct models coincides with the measured data under various operating conditions. It was found that the deviation between the two distinct models becomes more pronounced as the turbulent dissipation rate increases, which is opposite to the results in the numerical study of bubble columns. [ABSTRACT FROM AUTHOR]

• A novel design of tubular fission gas removal equipment with a serpentine pipe as the gas-liquid reactor. • The two-phase hydrodynamics in the vertical serpentine tube. • The gas removal performance of the novel device under various operating conditions. • A dimensionless correlation of modified Sherwood number. The separation of fission gases like 135-Xe and 85-Ke from the liquid nuclear fuel is a focused topic in the molten salt reactor (MSR) community. In this paper, a compact fission gas removal equipment with a serpentine tube as the gas-liquid contactor is introduced. The novel design is free of any moving or rotating components, bringing great convenience in the real-time fuel treatment. As crucial factors influencing the gas removal performance, the two-phase hydrodynamics and interfacial mass transfer characteristics in the vertical serpentine tube are investigated under a water-nitrogen-oxygen system. Both highspeed photographing and conductivity probe are employed to capture the evolution of bubbly flow inside the tubular contactor. The oxygen concentration differences between the inlet and outlet of the device are measured under various operating conditions, and the overall volumetric mass transfer coefficient k l a is derived to evaluate the gas removal efficiency. Experiment results imply a periodic transition of volume fraction profile between the off-center distribution in the downward flow and the near-wall distribution in the upward flow due to the centripetal effect in the U-bends. Bimodal distributions of bubble equivalent diameter are observed because of the intensive bubble coalescence inside the elbow. The overall k l a is found to be linearly proportional to the liquid Reynolds number and gas volume fraction. The dimensionless mass transfer correlation proposed in this work predicts well the overall volumetric mass transfer coefficient. [ABSTRACT FROM AUTHOR]

The study of bubbly flows relies on the extraction of bubble information in experiments. Extraction with image processing based on images taken by high-speed cameras is a commonly adopted approach. Current methods mostly deal with silhouettes, abandoning the grayscale information in the images. In this paper, we propose BubDepth, a workflow that utilizes grayscale information and automatically reconstructs rough 3D shapes of one side of the bubbles from single-view images. The workflow consists of two parts: segmentation and depth inference. A neural network is used to recognize bubbles and masks in the segmentation part. The following depth inference network computes a relative depth map for each mask, describing the 3D shapes of one side of bubbles. The neural networks are trained using a dataset generated by computer graphics techniques. The image generator can create synthetic images of scenes labeled with 3D shape information of bubbles. BubDepth is a novel method for the 3D reconstruction of bubble shape based on single-view images. It achieved accurate results for synthetic images and could produce convincing predictions in the tests for real images. • Reconstructs 3D shape of bubbles based on single-view images. • Generates realistic bubbly flow images with 3D shape information. • Bubble recognition and segmentation with neural networks. • The workflow achieved accurate results on the test dataset. [ABSTRACT FROM AUTHOR]