• Journal of the Korean Institute of Intelligent Systems
• /
• v.15 no.4
• /
• pp.437-442
• /
• 2005

The forecasting of flood runoff in the river is essential for flood control. The purpose of this study is to test a development of system for flood runoff forecasting using neural network model. For the flood events the tested rainfall and runoff data were the input to the input layer and the flood runoff data were used in the output layer To choose the forecasting model which would make up of runoff forecasting system properly, real-time runoff in the river when flood periods were forecasted by using the neural network model and the state-space model. A comparison of the results obtained by the two forecasting models indicated the superiority and reliability of the neural network model over the state-space model. The neural network model was modified to work in the Web and developed to be the basic model of the forecasting system for the flood runoff.

• 한국전산유체공학회:학술대회논문집
• /
• 2003.10a
• /
• pp.286-288
• /
• 2003

The physical model considered here is a horizontal layer of fluid heated below and cooled above with a periodic array of evenly spaced square cylinders placed at the center of the layer, whose aspect ratio here varies from unity to six. Periodic boundary condition is employed along the horizontal direction to allow for lateral freedom for the convection cells. Two-dimensional solution for unsteady natural convection is obtained using an accurate and efficient Chebyshev spectral multi-domain methodology for a given Rayleigh numbers of $10^6$

• Proceedings of the IEEK Conference
• /
• 2002.06d
• /
• pp.129-132
• /
• 2002

In this paper, we propose a 3-D animation system to track and analyze motion of the human object. The proposed system consists of two separate layers: motion analysis layer and 3-D model registration layer. Following the MPEG-4 SNHC standard, we generate object motion using body definition and animation parameters. In the implemented system, we acquire human motion data from a single camera and extract body definition parameters from arbitrary VRML human models.

• Journal of the Korean Chemical Society
• /
• v.57 no.3
• /
• pp.370-376
• /
• 2013

A back propagation artificial neural network model with one hidden layer is established to correlate the liquid-liquid equilibrium data of hydrocarbon-water systems. The model has four inputs and two outputs. The network is systematically trained with 48 data points in the range of 283.15 to 405.37K. Statistical analyses show that the optimised neural network model can yield excellent agreement with experimental data(the average absolute deviations equal to 0.037% and 0.0012% for the correlated mole fractions of hydrocarbon in two coexisting liquid phases respectively). The comparison in terms of average absolute deviation between the correlated mole fractions for each binary system and literature results indicates that the artificial neural network model gives far better results. This study also shows that artificial neural network model could be developed for the phase equilibria for a family of hydrocarbon-water binaries.

• 한국신재생에너지학회:학술대회논문집
• /
• 2005.06a
• /
• pp.697-700
• /
• 2005

Aquifer Thermal Energy Storage (ATES) system can be very cost-effective and renewable energy sources, depending on site-specific parameters and load characteristics. In order to develop an ATES system which has certain hydrogeological characteristics, understanding of the thermo hydraulic processes of an aquifer is necessary for a proper design of an aquifer heat storage system under given conditions. The thermo hydraulic transfer for heat storage is simulated using FEFLOW according to two sets of pumping and waste water reinjection scenarios of heat pump operation in a two layered confined aquifer. In the first set of model, the movement of the thermal front and groundwater level are simulated by changing the locations of injection and pumping well in seasonal cycle. However, in the second set of model the simulation is performed in the state of fixing the locations of pumping and injection well. After 365 days simulation period, the temperature distribution is dominated by injected water temperature and the distance from injection well. The small temperature change is appears on the surface compared to other slices of depth because the first layer has very low porosity and the transfer of thermal energy are sensitive at the porosity of each layer. The groundwater levels and temperature changes in injection and pumping wells are monitored to validate the effectiveness of the used heat pump operation method and the thermal interference between wells is analyzed.

• Journal of the Korean Vacuum Society
• /
• v.6 no.S1
• /
• pp.43-46
• /
• 1997

High resolution x-ray diffraction and two dimensional triple axis mapping were used to characterize a group of GaN layers of about 1.1$\mu$m grown by direct current plasma molecular beam epitaxy technique on 6H-SiC(0001). A FWHM of 11.9 arcmins for an $\omega$ scan and 1.2 arcmins for an $\omega$/2$\theta$ scan were observed. A careful study of the rocking curves showed there were some large mosaics in the GaN layer and a tilt of $0.029^{\circ}$ between the GaN layer and the SIC substrate was detected. The two dimensional triple axis mapping showed that the GaN mosaica were disoriented in the (0001) plane but rather uniformed in direction perpendicular to the plane. A mosaics were disoriented in the (0001) plane but rather uniformed in direction perpendicular to the plane. A mosaic model was deduced to explain the phenomenon and the dislocation density was estimated to be about~$10^9\;\textrm{cm}^{-2}$ acc ding to the model.

• Journal of Environmental Science International
• /
• v.13 no.7
• /
• pp.611-618
• /
• 2004

The Neural Network Models which mathematically interpret human thought processes were applied to resolve the uncertainty of model parameters and to increase the model's output for the streamflow forecast model. In order to test and verify the flood discharge forecast model eight flood events observed at Kumho station located on the midstream of Kumho river were chosen. Six events of them were used as test data and two events for verification. In order to make an analysis the Levengerg-Marquart method was used to estimate the best parameter for the Neural Network model. The structure of the model was composed of five types of models by varying the number of hidden layers and the number of nodes of hidden layers. Moreover, a logarithmic-sigmoid varying function was used in first and second hidden layers, and a linear function was used for the output. As a result of applying Neural Networks models for the five models, the N10-6model was considered suitable when there is one hidden layer, and the Nl0-9-5model when there are two hidden layers. In addition, when all the Neural Network models were reviewed, the Nl0-9-5model, which has two hidden layers, gave the most preferable results in an actual hydro-event.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.13 no.5
• /
• pp.999-1011
• /
• 1989

A consideration on the trubulence models for describing the redeveloping turbulent boundary layer beyond separation-reattachment in the flow over a backward-facing step is given through experimental and numerical studies. By considering the blance among the measured values of respective terms in the transport equations for the turbulent kinetic energy and the turbulent shear stress, the recovering process of the redeveloping boundary layer from non-equilibrium to equilibrium has been investigated, which takes place slowly over a substantial distance in the downstream direction. In the numerical study, the standard K-.epsilon. model and the Reynolds stress model have been applied to two kinds of flow regions, one for the entire downstream region after the backward-facing step and another for the downstream region after reattachment. Then the results are compared to a meaningful extent, with the experimental values of the turbulent kinetic energy k, the turbulent energy production term P, the dissipation term K-.epsilon. model, a necessity for a new modelling has been brought forward, which can be also applied to the case of the nonequlibrium turbulent flow.

• Journal of Environmental Science International
• /
• v.23 no.7
• /
• pp.1223-1232
• /
• 2014

Density currents have been easily observed in environmental flows, for instance turbidity currents and pollutant plumes in the oceans and rivers. In this study, we explored the propagation dynamics of density currents using the FLOW-3D computational fluid dynamics code. The renormalization group (RNG) $k-{\varepsilon}$ scheme, a turbulence numerical technique, is employed in a Reynold-averaged Navier-Stokes framework (RANS). The numerical simulations focused on two different types of intrusive density flows: (1) propagating into a two-layer ambient fluid; (2) propagating into a linearly stratified fluid. In the study of intrusive density flows into a two-layer ambient fluid, intrusive speeds were compared with laboratory experiments and analytical solutions. The numerical model shows good quantitative agreement for predicting propagation speed of the density currents. We also numerically reproduced the effect of the ratio of current depth to the overall depth of fluid. The numerical model provided excellent agreement with the analytical values. It was also clearly demonstrated that RNG $k-{\varepsilon}$ scheme within RANS framework is able to accurately simulate the dynamics of density currents. Simulations intruding into a continuously stratified fluid with the various buoyancy frequencies are carried out. These simulations demonstrate that three different propagation patterns can be developed according to the value of $h_n/H$ : (1) underflows developed with $h_n/H=0$ ; (2) overflows developed when $h_n/H=1$ ; (3) intrusive interflow occurred with the condition of 0 < $h_n/H$ < 1.

• Journal of Korean Society for Atmospheric Environment
• /
• v.13 no.5
• /
• pp.383-396
• /
• 1997

Numerical prediction of the pollutant dispersion over a two-dimensional hilly terrain is presented. The dispersion model used in the present work is based on the gradient diffusion theory and the finite-volume method on a non-orthogonal boundary-fitted grid system. The numerical model is validated by comparing the results with the available experimental data for the flat-floor dispersion within a turbulent boundary-layer. The numerical error analysis is performed based on the guideline of Kasibhatla et al.(1988) for the elevated-source dispersion in the flat-floor boundary layer having a power-law velocity and linear eddy-diffusivity profile. The influences of the two-dimensional hilly terrain on the dispersion from a continuously released source are numerically investigated by changing the emission locations and heights. It is found that the distributions of ground-level concentration are strongly influenced by the source location and the emission height. Hence, the terrain amplification factor is greatly enhanced when the pollutant source is located within a flow separation region. Dispersion from a source of short duration is also simulated and the duration time of the pollutant is compared at several downstream locations on a hilly terrain. The results of the numerical prediction are applied to the evaluation of environmental impacts due to the automobile exhausts at the seashore highway with a parallel mountain range.