• Proceedings of the Korean Society of Medical Physics Conference
• /
• 2006.08a
• /
• pp.33-33
• /
• 2006

• Geomechanics and Engineering
• /
• v.17 no.4
• /
• pp.385-392
• /
• 2019

The present fractional-order plasticity models for granular soil are mainly established under the triaxial compression condition, due to its difficult in analytically solving the fractional differentiation of the third stress invariant, e.g., Lode's angle. To solve this problem, a three dimensional fractional-order elastoplastic model based on the transformed stress method, which does not rely on the analytical solution of the Lode's angle, is proposed. A nonassociated plastic flow rule is derived by conducting the fractional derivative of the yielding function with respect to the stress tensor in the transformed stress space. All the model parameters can be easily determined by using laboratory test. The performance of this 3D model is then verified by simulating multi series of true triaxial test results of rockfill.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.15 no.1
• /
• pp.262-273
• /
• 1991

In the present study, the steady, incompressible, isothermal, developing flow in a 90.deg. rectangular cross sectional strongly curved duct with aspect ratio 1:1.5 and Reynolds number of 9.4*10$^{4}$ has been investigated. Measurements of components of mean velocities, pressures, and corresponding components of the Reynolds stress tensor are obtained with a hot-wire anemometer and pitot tube. In general, flow in a curved duct is characterized by the secondary vortices which are driven mainly by centrifugal force-radial pressure gradient imbalance, and the stress field stabilizing effects near the convex wall and destablizing effects close to the concave wall. It was found that the secondary mean velocities attain values up to 39% of the bulk velocity and are largely responsible for the convections of Reynolds stress in the cross stream plane. Therefor upstream of the bend the Reynolds stress are low. Corresponding to the small boundary layer thickness. At successive planes, large values of Reynolds stress were observed near the concave surface and the side wall.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.28 no.6
• /
• pp.672-681
• /
• 2004

Direct numerical simulation was carried out to study the vortical structures of the flow around a wall-mounted cube in a channel at Re=1,000 and Re=3,500 based on cubic height and bulk mean velocity. The cubic obstacle is situated in the entrance region of the channel flow where the boundary layers are developing. Upstream of the obstacle, steady and unsteady laminar horseshoe vortex systems are observed at Re=1,000 and Re=3,500, respectively; the near-wake flow is turbulent in both cases. The flow separates at each leading sharp edge of the cube, and subsequent vortex roll-up is noticed in the corresponding free-shear layer. The vortex shedding from the upper leading edge (upper vortices) and that from the two lateral leading edges (lateral vortices) are both quasi-periodic and their frequencies are computed. The upper and lateral vortices further develop into hairpin and Λ vortices, respectively. A series of instantaneous contours of the second invariant of velocity gradient tensor helps us identify spatial and temporal behaviors of the vortices in detail. The results indicate that the length and time scales of the vortical structures at Re=3,500 are much shorter than those at Re:1,000. Correlations between the upper and lateral vortices are also reported.

• Journal of Korea Water Resources Association
• /
• v.53 no.2
• /
• pp.97-105
• /
• 2020

Recently, Artificial Neural Network receives attention as a data prediction method. Among these, a Long Shot-term Memory (LSTM) model specialized for time-series data prediction was utilized as a prediction method of hydrological time series data. In this study, the LSTM model was constructed utilizing deep running open source library TensorFlow which provided by Google, to predict inflows of multipurpose dams. We predicted the inflow of the Yongdam Multipurpose Dam which is located in the upper stream of the Geumgang. The hourly flow data of Yongdam Dam from 2006 to 2018 provided by WAMIS was used as the analysis data. Predictive analysis was performed under various of variable condition in order to compare and analyze the prediction accuracy according to four learning parameters of the LSTM model. Root mean square error (RMSE), Mean absolute error (MAE) and Volume error (VE) were calculated and evaluated its accuracy through comparing the predicted and observed inflows. We found that all the models had lower accuracy at high inflow rate and hourly precipitation data (2006~2018) of Yongdam Dam utilized as additional input variables to solve this problem. When the data of rainfall and inflow were utilized together, it was found that the accuracy of the prediction for the high flow rate is improved.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.23 no.8
• /
• pp.1063-1071
• /
• 1999

Fine grid calculations are reported for the developing turbulent flow in a curved duct of square cross-section with a radius of curvature to hydraulic diameter ratio ${\delta}=Rc/D_H=3.357 $ and a bend angle of 720 deg. A sequence of modeling refinements is introduced; the replacement of wall function by a fine mesh across the sublayer and a low Reynolds number algebraic second moment closure up to the near wall sublayer in which the non-linear return to isotropy model and the cubic-quasi-isotropy model for the pressure strain are adopted; and the introduction of a multiple source model for the exact dissipation rate equation. Each refinement is shown to lead to an appreciable improvement in the agreement between measurement and computation.

• Journal of Ocean Engineering and Technology
• /
• v.31 no.3
• /
• pp.189-195
• /
• 2017

A three-dimensional higher order boundary element method based on the B-spline is presented. The method accurately models piecewise continuous bodies and induced velocity potentials using B-spline tensor product representations, and it is capable of obtaining accurate pointwise values for the potential and its derivatives, especially in the trailing edge and tip region of the lift generating body, which may be difficult or impossible to evaluate with constant panel methods. In addition, we implement a wake roll-up and examine the tip vortex formation in the near wake region. The results are compared with existing numerical results and the results of experiments performed out at the cavitation tunnel of Chungnam National University.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.20 no.8
• /
• pp.2593-2600
• /
• 1996

A modified k-$\varepsilon$ turbulence model having a generality is proposed in the present study, in which the constant $C_{\varepsilon2}$in the $\varepsilon$-equation is simply changed as a functional form of a new parameter both satisfying the tensor invariant condition and representing the extra straining effect on complex shear flows. With this model turbulent shear flows over two-dimensional obstacles placed in a channel are numerically studied for different blockage ratios and aspect ratios. Comparing with the available experimental data, the predicted results with the present model provide definite improvements over the standard model's results and work fairly well with the experimental data on the size of the recirculation zone, as well as mean velocity, wall static pressure, turbulent kinetic energy and Reynolds stresses.

• Annual Conference on Human and Language Technology
• /
• 2018.10a
• /
• pp.498-503
• /
• 2018

본 논문은 한국관광공사에서 제공하는 Tour API 3.0 Open API에서 제공하는 데이터를 바탕으로 한다. Google에서 제공해 주는 TensorFlow를 통해서 인공 신경망 딥러닝 알고리즘과 가중치 알고리즘을 통해서 사용자 기호에 맞춰 정보를 추천해 주는 어플리케이션 '눈치코칭_여행딥러닝'의 설계 및 구현에 대하여 서술한다. 현재 순위알고리즘은 평균적으로 40%, 딥러닝 모델은 60%정확도를 보여, 딥러닝이 보다 좋은 성능을 보였다.

• Journal of the Semiconductor & Display Technology
• /
• v.19 no.2
• /
• pp.100-104
• /
• 2020

Over the past several years, deep learning has been widely used for feature extraction in image and video for various applications such as object classification and facial recognition. This paper introduces an implantation of embedded Linux system for embossed digits recognition using CNN based deep learning methods. For this purpose, we implemented a coin recognition system based on deep learning with the Keras open source library on Raspberry PI. The performance evaluation has been made with the success rate of coin classification using the images captured with ultra-wide angle camera on Raspberry PI. The simulation result shows 98% of the success rate on average.