Begell House Inc.
Multiphase Science and Technology
MST
0276-1459
18
1
2006
INTERFACIAL CHARACTERISTICS OF TWO-PHASE FLOW
1-29
10.1615/MultScienTechn.v18.i1.10
Xiaodong
Sun
Nuclear Engineering Program, Department of Mechanical and Aerospace Engineering, The Ohio State University, 201 W 19th Avenue, Columbus, OH 43210, USA
Mamoru
Ishii
Therma-Hydraulics and Reactor Safety Laboratory, School of Nuclear Engineering, Purdue University, 400 Central Drive, West Lafayette, IN 47907, USA
This paper presents new experimental and modeling approaches in characterizing interfacial structures in gas-liquid two-phase flow. For the experiments, two objective approaches are developed to identify flow regimes and to obtain local interfacial structure data. First, a global measurement technique using a non-intrusive ring-type impedance void-meter and a self-organizing neural network is presented to identify the “one-dimensional” flow regimes. In the application of this measurement technique, two methods are discussed, namely, one based on utilizing various statistical moments of the probability density function of the impedance probe's signal (PDF input method) and the other based on the sorted impedance signals, which is essentially the cumulative probability distribution function of the impedance signals (instantaneous direct signal input method). In the latter method, the identification can be made almost instantaneously since the required signals can be acquired over a very short time period. In addition, a double-sensor conductivity probe can also be used to obtain “local” flow regimes by using the instantaneous direct signal input method with the bubble chord length information. Furthermore, a newly designed conductivity probe with multiple double-sensor heads is proposed to obtain “two-dimensional” flow regimes across the flow channel. Secondly, a state-of-the-art four-sensor conductivity probe technique has been developed to obtain detailed local interfacial structure information. The four-sensor conductivity probe accommodates the double-sensor probe capability and can be applied in a wide range of flow regimes spanning from bubbly to churn-turbulent flows. The signal processing scheme is developed such that it categorizes the acquired parameters into two groups based on bubble cord length information. Furthermore, for the modeling of the interfacial structure characterization, the interfacial area transport equation proposed earlier has been studied to provide a dynamic and mechanistic prediction tool for two-phase flow analysis. Based on detailed modeling of the bubble interactions, one-group and two-group interfacial area transport equations have been developed.
VOID DIFFUSION COEFFICIENT IN TWO-PHASE VOID DRIFT FOR SEVERAL CHANNELS OF TWO- AND MULTI-SUBCHANNEL SYSTEMS
31-54
10.1615/MultScienTechn.v18.i1.20
Akimaro
Kawahara
Faculty of Advanced Science
and Technology, Kumamoto University, Chuo-Ku,
Kurokami 2-39-1, Kumamoto City, 860-8555, Japan
Michio
Sadatomi
Department or Advanced Mechanical System, Graduate School of Science and Technology, Kumamoto University, Kurokami 2-39-1, Chuo-Ku, Kumamoto City, 860-8555, Japan
K.
Kano
Dept. of Mechanical Engineering and Materials Science, Kumamoto University, Kumamoto, 860-8555, Japan
Y.
Sasaki
Dept. of Mechanical Engineering and Materials Science, Kumamoto University, Kurokami 2-39-1, Kumamoto City, 860-8555, Japan
H.
Kudo
Dept. of Mechanical Engineering and Materials Science, Kumamoto University, Kurokami 2-39-1, Kumamoto City, 860-8555, Japan
To improve a void drift model incorporated in a subchannel analysis code, experimental data have been obtained for vertical air-water two-phase flows in several test channels of two- and multi-subchannel systems. In order to know the effects of the lattice of fuel rods on the void drift, the channels made up of two subchannels (i.e., the two-subchannel system) surrounded by square lattice rods or triangle tight lattice rods were included in the experiment. In addition, to know the effects of the number of subchannels, as the multi-subchannel system the channel consisted of six subchannels simulating a square lattice BWR fuel rod bundle was included. In each test channel, the data have been collected on the axial redistributions of flow rates of both phases and void fraction in the respective subchannels. In order to study the effects of two-phase flow regimes or void fraction on the void drift, the flow regimes covered were slug, churn and annular flows with various combinations of air and water flow rates. By fitting the above data with Lahey et al.'s void settling model, we have determined a void diffusion coefficient in their model. The void diffusion coefficient data were compared between the respective test channels as well as Tapucu et al.'s data for a square lattice two-subchannel system. It was found that the void diffusion coefficient was much smaller in the tight lattice channel than the square lattice channels, i.e., a channel size effect. Furthermore, the void diffusion coefficient could be well correlated with the turbulent Peclet number using a subchannel geometry factor, regardless of both the gap clearance between subchannels and the number of subchannels. It was also found that the correlation depends on the rod lattice.
TURBULENT SWIRLING WATER FLOW WITH OIL DROPLETS
55-72
10.1615/MultScienTechn.v18.i1.30
Takuro
Kouda
Graduate student, Division of Mechanical and System Eng., Kyoto Institute of Technology, Kyoto, Japan
Yoshimichi
Hagiwara
Department of Mechanical and System Engineering, Graduate School of Science and Technology, Kyoto Institute of Technology Goshokaido-cho, Matsugasaki, Sakyo-ku, Kyoto 606-8585, Japan
We have carried out measurements of the flow field of swirling water flow with immiscible oil droplets in a horizontal pipe. The swirl is obtained with a stationary vane inside the pipe. The flow is visualized with tracer particles and the images of visualized flow are captured from the side and downstream-end of the pipe with a progressive-scan video camera. The velocity field is measured from the captured images with the velocity gradient tensor method. The experimental results show that the low-density droplets have relative motion to the water flow. The axial mean velocity decreases because the droplets injected into the flow have had lower velocity than the local flow velocity. The periodic pulsative wake flow of some droplets causes the increase in the turbulence intensity.
MODELING AND HYBRID SIMULATION OF BUBBLY FLOW
73-110
10.1615/MultScienTechn.v18.i1.40
K.
Sakoda
Kobe University, Rokkodai, Nada, Kobe 657-8501, Japan
Kosuke
Hayashi
Graduate School of Engineering, Kobe University, 1-1, Rokkodai, Nada, Kobe
657-8501, Japan
N.
Shimada
Sumitomo Chemical Co. Ltd., Niihama, Ehime 792-8521, Japan
Computational Multi-Fluid Dynamics, CMFD, has come to play an important role in practical engineering and fundamental researches on multiphase flows. Since a multiphase flow is often comprised of interdependent elementary phenomena with length scales ranging from mesoscale to macroscale, a multipurpose CMFD should cover a wide range of scales by the combination of different numerical methods such as interface tracking, bubble tracking and averaging methods. This report reviews recent activities on CMFD for bubbly flows, in particular, an improvement of an interface tracking method based on volume tracking, two types of hybrid CMFD methods based on a combination of interface tracking and bubble tracking methods and that of interface tracking and averaging methods, and work in progress heading toward the hybrid combination of the three methods.