• Journal of Power System Engineering
• /
• v.18 no.3
• /
• pp.59-65
• /
• 2014

The computational fluid dynamics was carried out to evaluate turbulent models on the swirling flow of a gun-type gas burner(GTGB) according to the mesh size. The commercial SC/Tetra software was used for a steady-state, incompressible and three-dimensional numerical analysis. In consequence, the velocity magnitude from the exit of a GTGB and the flowrate predicted by the turbulent models of MP k-${\varepsilon}$, Realizable k-${\varepsilon}$ and RNG k-${\varepsilon}$ agree with the results measured by an experiment very well. Moreover, the turbulent kinetic energy predicted by the turbulent model of standard k-${\varepsilon}$ with mesh type C only agrees with the experimental result very well along the radial distance. On the other hand, the detailed prediction of the information of swirling flow field near the exit of a GTGB at least needs a CFD analysis using a fairly large-sized mesh such as a mesh type C.

• Journal of Bio-Environment Control
• /
• v.3 no.1
• /
• pp.10-19
• /
• 1994

Wind load is known to be one of major forces to influence the stability of agricultural structures. General flow fields were calculated to determine flow characteristics over the envelop of the following three types of greenhouses with arched roof : single span, twin span greenhouses, and two single span greenhouses apart 3m inbetween. Pressure coefficients along the envelop of greenhouse were numerically calculated by the k-$\varepsilon$ turbulence model, which lead to determine wind forces on it. Curvilinear coordinate for an arched roof and the upwind scheme were adopted for the study. The calculated pressure coefficients were validated with the avaliable data of Japanese Standard and NGAM Standard. The Magnitude of calculated forces over the envelop was not in good accordance with data except the windward wall. Even tile data of Japanese and NGAM Standard for validation deviated a lot from each other in quantity and quality. Such discrepancy may be attributed to different geometric and/or flow configuration conditions for experiments, or the insenstivity of the k-$\varepsilon$ turbulence model to recirculation flow.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2002.12a
• /
• pp.162-165
• /
• 2002

Q-learning은 강화학습의 한 방법으로서, 여러 분야에 널리 응용되고 있는 기법이다. 최근에는 Linear Quadratic Regulation(이하 LQR) 문제에 성공적으로 적용된 바 있는데, 특히, 시스템모델의 파라미터에 대한 구체적인 정보가 없는 상태에서 적절한 입력과 출력만을 가지고 학습을 통해 문제를 해결할 수 있어서 상황에 따라서 매우 실용적인 대안이 될 수 있다. Neural Q-learning은 이러한 Q-learning의 Q-value를 MLP(multilayer perceptron) 신경망의 출력으로 대치시킴으로써, 비선형 시스템의 최적제어 문제를 다룰 수 있게 한 방법이다. 그러나, Neural Q방식은 신경망의 구조를 먼저 결정한 후 역전파 알고리즘을 이용하여 학습하는 절차를 취하기 때문에, 시행착오를 통하여 신경망 구조를 결정해야 한다는 점, 역전파 알고리즘의 적용으로 인해 신경망의 연결강도 값들이 지역적 최적해로 수렴한다는 점등의 문제점을 상속받는 한계가 있다. 따라서, 본 논문에서는 Neural-0 학습의 도구로, 역전파 알고리즘으로 학습되는 MLP 신경망을 사용하는 대신 최근 들어 여러 분야에서 그 성능을 인정받고 있는 서포트 벡터 학습법을 사용하는 방법을 택하여, $\varepsilon$-SVR(Epsilon Support Vector Regression)을 이용한 Q-value 근사 기법을 제안하고 관련 수식을 유도하였다. 그리고, 모의 실험을 통하여, 제안된 서포트 벡터학습 기반 Neural-Q 방법의 적용 가능성을 알아보았다.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.32 no.5
• /
• pp.58-64
• /
• 2004

A new method to obtain approximate solutions by placing the only poles of the slow subsystem for the two-time-scale aircraft dynamic systems. The two kinds of approximate solutions are obtained by a matrix block diagonalization. One is called the uncorrected solution, and the other is called the corrected solution. The former has an error of $O({\varepsilon})$, and the latter has an error of $O({\varepsilon}^2)$. Of course, both solutions are robust enough even though they are reduced solutions. The excellence of the proposed method is illustrated by an numerical example of an aircraft longitudinal dynamics.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.24 no.2
• /
• pp.204-213
• /
• 2000

This study deals with jet impingement, which is extensively used in the process industries to achieve intense heating, cooling or drying rates and also widely employed as a test flow for turbulent models due to its complex flow configuration, on a flat plate by numerical methods. In this calculation, the finite volume method was employed to solve the Navier-stokes equation based on the non-orthogonal coordinate with non-staggered variable arrangement. To get a better understanding for the fluid flow and heat transfer characteristics of the turbulent jet impingements, $k-{\varepsilon}-{\overline{v^{'2}}}$ turbulent model was adapted and compared with the experimental data and the result of standard $k-{\varepsilon}$ turbulent model. Numerical calculations were carried out with various flow rates, nozzle to plate distances. In the case of the axisymmetric jet impingement on a flat plate, $k-{\varepsilon}-{\overline{v^{'2}}}$ turbulent model showed better agreement with the experimental data than the standard $k-{\varepsilon}$ turbulent model in the prediction of the mean velocity profiles, the turbulent velocity profiles. the turbulent shear stress and the heat transfer rate. The highest heat transfer rate can be obtained when the impingement occurs within the potential core..

• Nuclear Engineering and Technology
• /
• v.5 no.1
• /
• pp.30-37
• /
• 1973

The Nishitani's equations for impedance of anodic oxide films have been derived based on a p-i-n model under the assumption of $\omega$$\varepsilon$$\rho$$_{ο}$<<4$\pi$<<$\omega$$\varepsilon$$\rho$$_{\omega}$, where $\omega$ is angular frequency, $\varepsilon$ is dielectric constant, and $\rho$$_{ο}$ and $\rho$$_{\omega}$ are the resistivity of the interface region and the intrisic region of the anodic oxide film, respectively. Since it is not possible to evaluate all parameters in the equations, however, any clear physical picture cannot be obtained from the equations. Therefore, the equations are modified under the assumption of $\omega$$\tau$$_{\omega}$>>1 and In(1+$\omega$$^2$$\tau$$_{ο}$$^2$)<<1, where $\tau$$_{\omega}$=$\varepsilon$$\rho$$_{\omega}$(4$\pi$) and $\tau$$_{ο}$=$\varepsilon$$\rho$$_{ο}$/(4$\pi$). The modified equations are then used to explain the change in the frequency characteristics of anodic oxide films when they are heated. The change in impedance of anodic oxide films when they are heated is attributed mainly to the increase in the diffusion layer and to the decrease in the resistivity of anodic oxide films.s.

• Journal of the Korean Society of Propulsion Engineers
• /
• v.9 no.3
• /
• pp.49-59
• /
• 2005

Turbulent heat transfer over a fully-developed wavy channel is investigated by a turbulence model. The nonlinear k- f - f$_{ model of Park et at.[1] is slightly modified and their explicit algebraic heat flux model is employed. The Reynolds number is fixed at Re$_{b}$=6760 and the wave configuration is varied in the range of 0 $\leq$ $\alpha$/$\lambda$$\leq$0.15 and 0.25 $\leq$A/H$\leq$4.0. In order to verify model performances, a large eddy simulation is performed for the selected cases. The model performance is shown to be generally satisfactory. By using k- $\varepsilon$ - f$_{ model, the enhancement of heat transfer and the characteristics of turbulent flow in wavy wall are investigated. Finally, the influence of wavy configuration on heat transfer is scrutinized.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.36 no.2
• /
• pp.189-196
• /
• 2012

The heavy moderator acts as the ultimate heat-sink in an operating CANDU reactor. HUKINS has been developed to investigate moderator flow patterns. HUKINS consists of a 38.4-mm-thick cylindrical shell with a 0.95 m inner diameter and 88 sus-tubes that produce a total heat of 10 kW. A chemical visualization method was selected to estimate the occurrence of typical moderator flow patterns. Momentum-dominated flow, mixed flow, and buoyancy-dominated flow are detected under conditions of a heat load of 7.7 kW and input mass flow rates of 4, 7, and 11 L/min. The experimental results are similar to the results of a CFD simulation that consisted of approximately 1.9 million grids and was conducted using the k-${\varepsilon}$ turbulence model. Therefore, both the present experiments and simulations using HUKINS, a 1/8-scale model, represent all three important flow patterns expected in the real CANDU6 reference reactor. Thus, it has been demonstrated that HUKINS could be useful in the study of CANDU6 moderator circulation.

• Journal of Energy Engineering
• /
• v.8 no.2
• /
• pp.333-340
• /
• 1999

A numerical study of turbulent condensing vapor jet submerged in subcooled liquids has been conducted. A physical model of the process is presented employing the locally homogeneous flow approximation of two phase flow in conjunction with a $\kappa$-$\varepsilon$-g model of turbulence properties. In this model the turbulence is represented by differential equations for its kinetic energy and dissipation. A differential equation for the concentration fluctuations is solved and a clipped normal probability distribution function is proposed for the mixture fraction. Effects of steam mass flux, pool temperature and nozzle internal diameter on the condensing vapor jet are also analyzed. The model is evaluated using existing data for turbulent condensing vapor jets. The agreement between the predictions and the available experimental data is good.

• Nuclear Engineering and Technology
• /
• v.26 no.3
• /
• pp.345-354
• /
• 1994

Numerical Predictions including secondary flows have been Performed for fully developed turbulent single-phase rod bundle flows. The k-$\varepsilon$ turbulence model(two equation model) for the isotropic eddy viscosity, together with an algebraic stress model for generating secondary velocities, enabled the prediction of mean axial velocities, secondary velocities, and turbulent kinetic energy and turbulent stresses. Comparisons with experiment hate shown that the influence of secondary motion on mean flow and turbulence is dearly evident. The convective transport effects of secondary flow on the velocity field have been identified.