[en] Recently, efficient heat dissipation has become necessary because of the miniaturization of devices, and research on boiling on micro-channels has attracted attention. However, in the case of micro-channels, the friction coefficient and heat transfer characteristics are different from those in macro-channels. This leads to large errors in the micro scale results, when compared to correlations derived from the macro scale. In addition, due to the complexity of the mechanism, the boiling phenomenon in micro-channels cannot be approached only by experimental and theoretical methods. Therefore, numerical methods should be utilized as well, to supplement these methods. However, most numerical studies have been conducted on macro-channels. In this study, we applied the lattice Boltzmann method, proposed as an alternative numerical tool to simulate the boiling phenomenon in the micro-channel, and predicted the bubble growth process in the channel.

[en] For numerical analysis of multiphase flow, each interface boundary should be captured, and the geometric deformation of the interface needs to be predicted. To predict the interface, the singular interface model and diffusion interface model can be used. Among them, free energy based lattice Boltzmann method has adopted the diffusion interface model, with which it is easy to simulate complex multiphase flow phenomena such as bubble collapse, droplet collision, and moving contact lines. A new lattice Boltzmann method for the simulation of multiphase flows is described, and test results for the validation are presented. Finally, some simulations were carried out for the investigation of dynamic behavior of multiple rising bubbles.

[en] In the present study, temperature evaluations on long-term storage of radioactive waste produced in the process of isotope production were performed using two different methods. Three-dimensional analysis was carried out assuming a volumetric heat source, while two-dimensional studies were performed assuming a point source. The maximum temperature difference between the predictions of the volumetric and point source models was approximately 5°C. For the conceptual design level, a point source model may be suitable to obtain the overall temperature characteristics of different loading locations. For more detailed analysis, the model with the volumetric source may be applicable to optimize the loading pattern in order to obtain minimum temperatures.

[en] Generated neutrons are transferred through the beam-tubes of a research reactor. When a beam-tube rupture occurs inside the pool, the pool water enters quickly into the tube, and hits the flange foil at the other end. Unless the foil endures the impact pressure, a multiple beam-tube-rupture accident will occur. Therefore, to determine the thickness of the end flange foil, the impact pressure has to be estimated. In this study, the maximum impact pressure is estimated analytically by considering the water hammer phenomenon. The result is then validated with computational fluid dynamic (CFD) simulations. For the simulations, a commercially available CFD code, ANSYS CFX, is used. The analytic solutions show good agreement with the CFD results. (author)

[en] A decay tank shall be designed to provide enough flow residence time to ensure that the N-16 activity decreases before the coolant leaves the decay tank's shielding room. However, when a proper criterion for the flow residence time in a decay tank is not presented, the tank would be oversized/undersized. In this paper, design evaluation for a decay tank is performed by investigating the effect of the fluid distribution along the residence time on the total dose rate and the required minimum flow residence time. The evaluation is also carried out to resize the predesigned decay tank. (author)

[en] Highlights: ► A decay tank, which consists of perforated plates and cylinder, is presented. ► The fluid residence time in the decay tank is estimated. ► It is found that unsteady simulation is much more efficient than steady simulation for UDS method. -- Abstract: A decay tank shall be designed to provide enough flow residence time to ensure that the N-16 activity decreases before the coolant leaves the decay tank's shielding room. Therefore, estimation of the residence time in a decay tank is necessary before the conclusion of such a design. To estimate the flow residence time in a decay tank, the transport equation of a user-defined additional variable, which has a time unit and zero kinematic diffusivity, is solved using steady and unsteady assumptions. The minimum flow residence time is compared with that of an analytic method with an assumption of a uniform flow in a decay tank, and a particle tracking method. For the simulations, a commercially available CFD code, ANSYS-CFX, is used. When comparing the results, the unsteady simulation is seen to be much more efficient approach than steady simulation to obtain the minimum flow residence time of the decay tank

[en] Highlights: • Flow in fibrous porous media is studied using lattice Boltzmann method. • Multiple-relaxation-time (MRT) LBM scheme is used. • The structures of overlapping fibers yield 2.5 times larger permeability. • Fiber arrangement affects the permeability only when slip flow occurs. -- Abstract: The permeabilities of microscale fibrous porous media were calculated using the multiple-relaxation-time (MRT) lattice Boltzmann method (LBM). Two models of the microscale fibrous porous media were constructed based on overlapping fibers (simple cubic, body-centered cubic). Arranging the fibers in skew positions yielded two additional models comprising non-overlapping fibers (skewed simple cubic, skewed body-centered cubic). As the fiber diameter increased, the fibers acted as granular inclusions. The effects of the overlapping fibers on the media permeability were investigated. The overlapping fibers yielded permeability values that were a factor of 2.5 larger than those obtained from non-overlapping fibers, but the effects of the fiber arrangement were negligible. Two correlations were obtained for the overlapping and non-overlapping fiber models, respectively. The effects of the rarefaction and slip flow are also discussed. As the Knudsen number increased, the dimensionless permeability increased; however, the increase differed depending on the fiber arrangement. In the slip flow regime, the fiber arrangement inside the porous media became an important factor

[en] Highlights: ► We performed an experiment and numerical simulation to design the siphon break line size. ► The experimental loop was manufactured at a similar scale of a common research reactor. ► The CFD code employing turbulent and two-phase models was used to compare the experimental results. ► Siphon break phenomena are dependent on the air- and water-flow rates and a void fraction. -- Abstract: For guaranteeing the pool water inventory, which is important to the nuclear safety of a research reactor, a siphon breaker is installed to limit the pool water drain during and after all postulated initiating events in the research reactor. Because the main pipe size of the reactor is relatively large, the size of the siphon break line should be determined to break the siphon phenomena. The siphon breaker design is validated through experiments and numerical simulations. An experimental loop was manufactured at a similar scale of a common research reactor, and a commercially available CFD code was used to compare the experimental results. The undershooting height was measured with a camera and absolute pressure transducer according to the siphon break line sizes. The pressure and superficial velocity inside the main pipe according to the pool water level were analyzed to understand the siphon break phenomena. The CFD code was tested to determine its usefulness for simulating siphon break phenomena over the same conditions with the experiment using several models for the two-phase flow phenomena. The undershooting height, pressure, and liquid superficial velocity were calculated using homogeneous and inhomogeneous models with the SST turbulent model and compared with the experimental results. Although the results of the ANSYS CFD model show some differences with the experimental data, the CFD results using the inhomogeneous model show good agreement with the experimental data. In addition, the homogeneous model results can be used conservatively in the design of a siphon breaker.