• Korean Chemical Engineering Research
• /
• v.50 no.1
• /
• pp.61-65
• /
• 2012

The onset of convection in a mushy layer is analyzed by using linear stability theory in time-dependent solidification of a binary melt. A simplified model of a near-eutectic mush, in which the mush is assumed to be a porous block, is used and the propagation theory is applied to determine the critical conditions for the onset of convection. The present critical Rayleigh number is higher than the existing experimental result and also theoretical results obtained by considering the mushy layer with an overlying liquid layer. The constant pressure (permeable) condition applied on the mush-liquid interface produces a lower critical Rayleigh number, which is closer to the experimental results of aqueous ammonium chloride solution, compared with the impermeable condition.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.22 no.6
• /
• pp.771-787
• /
• 1998

A Navier-Stokes code with a modified low Reynolds number k-.epsilon. turbulence model was used to study the unsteady transitional boundary layer flow due to rotor-stator interaction. The modification, proposed by Launder, to improve prediction of stagnation flows was incorporated to the low Reynolds number k-.epsilon. turbulence model by Fan-Lakshminarayana-Barnett. Numerical solution is shown to capture well the calmed laminar flow as well as the wake induced transitional strip due to rotor-stator interaction and shows improvement, in terms of onset of transition and its length, over previous Euler/boundary layer solution. The turbulent kinetic energy shows local maximum along the upstream rotor wake in the wake induced transitional strip and this characteristics is observed untill the end of transition. The wake induced strip also shown apparent even in the laminar sublayer as the upstream rotor wake penetrates inside the boundary layer.

• 한국해양학회지
• /
• v.28 no.1
• /
• pp.17-23
• /
• 1993

A right cylindrical tank with sloping bottom and top (${\beta}-effect$) is filled with two-layered fluid and is put on the rotating table. External fluid of same density as the lower-layer fluid is continuously injected to drive the lower-layer current. By minimizing the interfacial stress between two layers the motion in the lower-layer deformed the shape of interface such that the upper-layer adjust itself to the variations of the interface in terms of its direction of flow patterns .The most significant parameter is the internal Froude Number($F_1$) and when $F_1$ is greater than 6 two-cellular circulation of the upper-layer changes its direction, there by creates a separation of Western boundary current. The separation position moves to the most northward when $F_1$ equals to 6.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.14 no.2
• /
• pp.216-221
• /
• 2004

This paper describes an optimal synthesis method of binary neural network for pattern recognition. Our objective is to minimize the number of connections and the number of neurons in hidden layer by using a Newly Expanded and Truncated Learning Algorithm (NETLA) for the multilayered neural networks. The synthesis method in NETLA uses the Expanded Sum of Product (ESP) of the boolean expressions and is based on the multilayer perceptron. It has an ability to optimize a given binary neural network in the binary space without any iterative learning as the conventional Error Back Propagation (EBP) algorithm. Furthermore, NETLA can reduce the number of the required neurons in hidden layer and the number of connections. Therefore, this learning algorithm can speed up training for the pattern recognition problems. The superiority of NETLA to other learning algorithms is demonstrated by an practical application to the approximation problem of a circular region.

• 유체기계공업학회:학술대회논문집
• /
• 2006.08a
• /
• pp.191-194
• /
• 2006

In this paper, we present a detailed mechanism of drag reduction by dimples and roughness on a sphere by measuring the streamwise velocity above the dimpled and roughened surfaces, respectively. Dimples cause local flow separation and trigger the shear layer instability along the separating shear layer, resulting in generation of large turbulence intensity. With this increased turbulence, the flow reattaches to the sphere surface with high momentum near the wall and overcomes strong adverse pressure gradient formed in the rear sphere surface. As a result, dimples delay main separation and reduce drag significantly. The present study suggests that generation of a separation bubble, i.e. a closed-loop streamline consisting of separation and reattachment, on a body surface is an important flow-control strategy for drag reduction on a bluff body such as the sphere and cylinder. In the case of roughened sphere, the boundary layer flow is directly triggered by roughness and changes to a turbulent flow. Due to this change, the drag significantly decreases. As the Reynolds number further increases, transition to turbulence occurs earlier on the sphere surface. Because of faster growth of turbulent boundary layer by roughness, earlier transition thickens the boundary layer, resulting in earlier separation and drag increase with increasing Reynolds number

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• v.2
• /
• pp.101-106
• /
• 2006

The geoidal undulations are needed for determining the orthometric heights from the Global Positioning System GPS-derived ellipsoidal heights. There are several methods for geoidal undulation determination. The paper presents a method employing a simple architecture Two Layer Multiquadric-Biharmonic Artificial Neural Network (TLMB-ANN) to approximate an area of 4200 square kilometres quasigeoid surface with GPS-levelling data. Hardy’s Multiquadric-Biharmonic functions is used as the hidden layer neurons’ activation function and Levenberg-Marquardt algorithm is used to train the artificial neural network. In numerical examples five surfaces were compared: the gravimetric geometry hybrid quasigeoid, Support Vector Machine (SVM) model, Hybrid Fuzzy Neural Network (HFNN) model, Traditional Three Layer Artificial Neural Network (ANN) with tanh activation function and TLMB-ANN surface approximation. The effectiveness of TLMB-ANN surface approximation depends on the number of control points. If the number of well-distributed control points is sufficiently large, the results are similar with those obtained by gravity and geometry hybrid method. Importantly, TLMB-ANN surface approximation model possesses good extrapolation performance with high precision.

• Journal of Energy Engineering
• /
• v.24 no.3
• /
• pp.135-141
• /
• 2015

The core melts stratifies into lower mixture layer and upper metal layer by density in a severe accident condition. The decay heat generated from the mixture layer threatens the integrity of the reactor vessel. This study simulated the natural convection heat transfer of the mixture layer with volumetric heat source using the mass transfer system. $H_2SO_4-CuSO_4$ electroplating system was used as the mass transfer system. With the modified Rayleigh number of $3{\times}10^{14}$, the Nusselt number showed minimum at the bottom and increased along curvature to the top of the experimental apparatus.

• Korean Journal of Air-Conditioning and Refrigeration Engineering
• /
• v.10 no.3
• /
• pp.260-270
• /
• 1998

The viscous plane stagnation-flow solidification problem is theoretically investigated. An analytic solution at the beginning of solidification is obtained by expanding the temperature and thickness of solidified layer in powers of time. An exact expression for the steady-state thickness of solidified layer is also obtained. The .fluid flow toward the cold substrate inhibits the solidification process. As Stefan number becomes larger, or Prandtl number becomes smaller, the solidification is more strongly inhibited by the fluid flow. The transient heat flux at the liquid side of solid-liquid interface is increased, as Stefan number or Prandtl number is increased.

• Proceedings of the Korean Institute of Surface Engineering Conference
• /
• 2017.05a
• /
• pp.97-97
• /
• 2017

This study investigated the corrosion protection by electrodeposited copper layer on AZ31 Mg alloy with and without additional silver layer by immersion test, salt spray test, OCP transient and potentiodynamic polarization experiment. The single electrodeposited Cu layer on AZ31 Mg alloy showed a nodular structure with many imperfections of crevices between the nodules, which resulted in the fast initiation of pitting corrosion within first few hours of immersion. Double-layer coating of Cu and outer Ag layer slightly increased the initiation time for pitting corrosion. Triple-layer coatings of Cu/Ag/Cu exhibited the most efficient corrosion protection of AZ31 Mg alloy, compared to the single- and double-layer coatings. Surface morphology of the outer Cu layer in the triple-layer was changed from the nodular structure to fine particle structure with no crevices due to the presence of an additional Ag layer. Thus, the improved corrosion resistance of AZ31 Mg alloy by electrodeposited Cu/Ag or Cu/Ag/Cu layers is readily ascribed to the decreased number of imperfections in the electrodeposited layers due to the additional silver layer. It is concluded that the additional silver layer provides many nucleation sites for the second Cu plating, resulting in the formation of finer and denser structure than the first Cu electrodeposit.

• Journal of computational fluids engineering
• /
• v.4 no.2
• /
• pp.17-30
• /
• 1999

The unsteady transitional boundary layer due to rotor-stator interaction was studied at two operation points, the design and one off design points. The off design point leads to lower blade loading and lower Reynolds number. A Navier-Stokes code developed in the previous study was parallelized to expedite computations. A low Reynolds number turbulence model was used to close the momentum equations. All computations show good agreement with experimental data. The wake induced transitional strip on the suction side of the stator is clearly captured at design point operation. There is no noticeable change in shape and phase angle of the wake induced strip even in the laminar sublayer. The wake induced transitional strip at off design point shows more complex structure. The wake induced transitional strip is observed only in the turbulent layer, and becomes obscure in the laminar sublayer and buffer layer. This behavior is probably consequent upon that the transition is governed by both wake induced strip and natural transition mechanism by Tollmien-Schlichting wave.