• Journal of Korean Society of Transportation
• /
• v.22 no.5
• /
• pp.123-137
• /
• 2004

The study on traffic assignment is actively being performed which reflect networks status using time. Its background is increasing social needs to use traffic assignment models in not only hardware area of road network plan but also software area of traffic management or control. In addition, multi-class traffic assignment model is receiving study in order to fill a gap between theory and practice of traffic assignment model. This model is made up of two, one of which is multi-driver class and the other multi-vehicle class. The latter is the more realistic because it can be combined with dynamic model. On this background, this study is to build multidynamic model combining the above-mentioned two areas. This has been a theoretic pillar of ITS in which dynamic user equilibrium assignment model is now made an issue, therefore more realistic dynamic model is expected to be built by combining it with multi-class model. In case of multi-vehicle, FIFO would be violated which is necessary to build the dynamic assignment model. This means that it is impossible to build multi-vehicle dynamic model with the existing dynamic assignment modelling method built under the conditions of FIFO. This study builds dynamic network model which could relieve the FIFO conditions. At the same time, simulation method, one of the existing network loading method, is modified to be applied to this study. Also, as a solution(algorithm) area, time dependent shortest path algorithm which has been modified from existing shortest path algorithm and the existing MSA modified algorithm are built. The convergence of the algorithm is examined which is built by calculating dynamic user equilibrium solution adopting the model and algorithm and grid network.

• The KIPS Transactions:PartA
• /
• v.13A no.1 s.98
• /
• pp.45-52
• /
• 2006

To perform on-line maintenance for Distributed Information System, it is indispensable to disseminate files to participant nodes in the network. When users' requests for file deployment occur simultaneously in a short period a deployment server falls into overload phase, which is often called Flash Crowds. h common solution to avoid Flash Crowds is to increase hardware capacity. In this paper, we propose a software solution based on P2P, which does not cost any additional expense. In the proposed solution, nodes in the network are grouped into subnetworks one of which is composed of only neighboring nodes. In each subnetwork, copies of deployment files can be transferred to each other. Consequently, it brings about the effect of load balancing in deployment server. To raise the effectiveness, target files for deployment are packed into one package. Before being transferred, each package is divided into multiple equal-sized segments. A deployment server in a normal phase transmits a package requested from nodes in segment units. However a deployment server is overloaded, if segments already exist in the subnetwork participant nodes in the subnetwork receive necessary segments from neighboring nodes. In this paper, we propose data structures and algorithm for this approach and show performance improvement through simulation.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.13 no.9
• /
• pp.3815-3821
• /
• 2012

The LCD frame is an important part which supports the BLU of medium/large sized TFT-LCD. To produce it efficiently, it is necessary to achieve the molding process improvement from 1 cavity to 2 cavity system. Because 2 cavity mold is compact and its hot-runner zone is broadened, it is difficult to control the temperature on the mold. In this study, injection molding analysis on the frame in 2 cavity process with FEA(Finite Element Analysis) software is carried out to estimate its quality. The calculated injection molding pressures and maximum deflection in 1 and 2 cavity processes are 41.13 MPa and 1.62 mm, 40.49 MPa and 1.66 mm respectively. The measured maximum flexure load and surface roughness of the left and right frame of 2 cavities are 209 N and 0.08 ${\mu}m$, 193 N and 0.10 ${\mu}m$ while those in 1 cavity are 140 N and 0.13 ${\mu}m$. Thermal image shows that the maximum standard deviation of the temperature on left and right side of 2 cavity mold is $1.23^{\circ}C$. The simulation and measurement results show that the quality of the frame in 2 cavity injection molding process as a whole is not worse than that of 1 cavity system. But maximum flexure loads of the frame in 2 cavity process are far greater than that in 1 cavity process.

• Journal of Electrical Engineering and Technology
• /
• v.8 no.6
• /
• pp.1474-1480
• /
• 2013

Silicon carbide (SiC)-zirconium diboride ($ZrB_2$) composites were prepared by subjecting a 60:40 vol% mixture of ${\beta}$-SiC powder and $ZrB_2$ matrix to spark plasma sintering (SPS) in 15 $mm{\Phi}$ and 20 $mm{\Phi}$ molds. The 15 $mm{\Phi}$ and 20 $mm{\Phi}$ compacts were sintered for 60 sec at $1500^{\circ}C$ under a uniaxial pressure of 50 MPa and argon atmosphere. Similar composites were simulated using $Flux^{(R)}$ 3D computer simulation software. The current and power densities of the specimen sections of the simulated SiC-$ZrB_2$ composites were higher than those of the mold sections of the 15 $mm{\Phi}$ and 20 $mm{\Phi}$ mold simulated specimens. Toward the centers of the specimen sections, the current densities in the simulated SiC-$ZrB_2$ composites increased. The power density patterns of the specimen sections of the simulated SiC-$ZrB_2$ composites were nearly identical to their current density patterns. The current densities of the 15 $mm{\Phi}$ mold of the simulated SiC-$ZrB_2$ composites were higher than those of the 20 $mm{\Phi}$ mold in the center of the specimen section. The volume electrical resistivity of the simulated SiC-$ZrB_2$ composite was about 7.72 times lower than those of the graphite mold and the punch section. The power density, 1.4604 $GW/m^3$, of the 15 $mm{\Phi}$ mold of the simulated SiC-$ZrB_2$ composite was higher than that of the 20 $mm{\Phi}$ mold, 1.3832 $GW/m^3$. The $ZrB_2$ distributions in the 20 $mm{\Phi}$ mold in the sintered SiC-$ZrB_2$ composites were more uniform than those of the 15 $mm{\Phi}$ mold on the basis of energy-dispersive spectroscopy (EDS) mapping. The volume electrical resistivity of the 20 $mm{\Phi}$ mold of the sintered SiC-$ZrB_2$ composite, $6.17{\times}10^{-4}{\Omega}cm$, was lower than that of the 15 $mm{\Phi}$ mold, $9.37{\times}10^{-4}{\Omega}{\cdot}cm$, at room temperature.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.38 no.9
• /
• pp.1029-1036
• /
• 2014

Several types of automatic 3D scanners are available for use in the 3D scanning industry, e.g., an automatic 3D scanner that uses a robot arm and one that uses an automatic rotary table. Specifically, these scanners are used to obtain a 3D shape using automatic assisting devices. Most of these scanners are required to perform numerous operations, such as merging, aligning, trimming, and filling holes. We are interested in developing an automatic 3D shape collection device using a spherical-coordinate-based guiding system. Then, the aim of the present study is to design an automatic guiding system that can automatically collect 3D shape data. We develop a 3D model of this system and measuring data which are collected by a personal computer. An optimal design of this system and the geometrical accuracy of the measured data are both evaluated using 3D modeling software. The developed system is then applied to an object having a highly complex shape and manifold sections. Our simulation results demonstrate that the developed system collects higher-quality 3D data than the conventional method.

• Journal of Korean Society of Steel Construction
• /
• v.25 no.3
• /
• pp.307-320
• /
• 2013

This paper proposes a new hybrid support structure by the driven piles which removes disadvantages of the existing type of support structure for offshore wind turbines. The hybrid type of support structure is combined with concrete cone and steel shaft, and is supported not only by gravity type foundations but also by driven piles. For three dimensional analysis of the huge and thick concrete structure, a solid-shell element that is capable of exact modeling and node interpolations of stresses is developed. By applying wave theory of stream function and solid-shell element in XSEA simulation software for fixed offshore wind turbines, a quasi-static analysis and natural frequency analysis of proposed support structure are performed with the environmental condition on Southwest Coast in Korea. In the result, lateral displacement is not exceed allowable displacement and a superiority of dynamic behavior of new hybrid support structure is validated by natural frequency analysis. Consequently, the hybrid support structure presented in this study has a structural stability enough to be applied on real-site condition in Korea. The optimized structures based on the preliminary design concept resulted in an efficient structure, which reasonably reduces fabrication costs.

• Proceedings of the Korean Vacuum Society Conference
• /
• 2013.02a
• /
• pp.176-177
• /
• 2013

Surface plasmon polaritons (SPPs) have attracted the attention of scientists and engineers involved in a wide area of research, microscopy, diagnostics and sensing. SPPs are waves that propagate along the surface of a conductor, usually metals. These are essentially light waves that are trapped on the surface because of their interaction with the free electrons of conductor. In this interaction, the free electrons respond collectively by oscillating in resonance with the light wave. The resonant interaction between the surface charge oscillation and the electromagnetic field of the light constitutes the SPPs and gives rise to its unique properties. In this papers, we studied theoretical and experimental extraordinary transmittance (T) and reflectance (R) of 2 dimensional metal hole array (2D-MHA) on GaAs in consideration of the diffraction orders. The 2d-MHAs was fabricated using ultra-violet photolithography, electron-beam evaporation and standard lift-off process with pitches ranging from 1.8 to $3.2{\mu}m$ and diameter of half of pitch, and was deposited 5-nm thick layer of titanium (Ti) as an adhesion layer and 50-nm thick layer of gold (Au) on the semiinsulating GaAs substrate. We employed both the commercial software (CST Microwave Studio: Computer Simulation Technology GmbH, Darmstadt, Germany) based on a finite integration technique (FIT) and a rigorous coupled wave analysis (RCWA) to calculate transmittance and reflectance. The transmittance was measured at a normal incident, and the reflectance was measured at variable incident angle of range between $30^{\circ}{\sim}80^{\circ}$ with a Nicolet Fourier transmission infrared (FTIR) spectrometer with a KBr beam splitter and a MCT detector. For MHAs of pitch (P), the peaks ${\lambda}$ max in the normal incidence transmittance spectra can be indentified approximately from SP dispersion relation, that is frequency-dependent SP wave vector (ksp). Shown in Fig. 1 is the transmission of P=2.2 um sample at normal incidence. We attribute the observation to be a result of FTIR system may be able to collect the transmitted light with higher diffraction order than 0th order. This is confirmed by calculations: for the MHAs, diffraction efficiency in (0, 0) diffracted orders is lower than in the (${\pm}x$, ${\pm}y$) diffracted orders. To further investigate the result, we calculated the angular dependent transmission of P=2.2 um sample (Fig. 2). The incident angle varies from 30o to 70o with a 10o increment. We also found the splitting character on reflectance measurement. The splitting effect is considered a results of SPPs assisted diffraction process by oblique incidence.

• Journal of KIISE:Computing Practices and Letters
• /
• v.15 no.10
• /
• pp.724-739
• /
• 2009

This paper suggests the energy-efficient message delivery scheme and the software platform which exploits the multiple network interfaces of the mobile terminals and GPS in the current mobile devices. The mobile terminals determine the delivery method among 'direct', 'indirect', and 'WAN' based on the position information of itself and other terminals. 'Direct' method sends a message directly to the target terminal using local RAT. 'Indirect' method extends the service area by exploiting intermediate terminals as relay node. If the target terminal is too far to reach through 'direct' or 'indirect' method, the message is sent using wireless WAN technology. Our proposed scheme exploits the position information and, thus, power consumption is drastically reduced in determining handover time and direction. Network simulation results show that our proposed delivery scheme improves the message transfer efficiency and the handover detection latency. We implemented a message platform in a smart phone realizing the proposed delivery scheme. We compared our platform with other typical message platforms from energy efficiency aspect by observing the real power consumption and applying the mathematical modeling. The comparison results show that our platform requires significantly less power.

• Journal of KIISE:Software and Applications
• /
• v.30 no.3_4
• /
• pp.326-338
• /
• 2003

The objective of a neural network design and model selection is to construct an optimal network with a good generalization performance. However, training data include noises, and the number of training data is not sufficient, which results in the difference between the true probability distribution and the empirical one. The difference makes the teaming parameters to over-fit only to training data and to deviate from the true distribution of data, which is called the overfitting phenomenon. The overfilled neural network shows good approximations for the training data, but gives bad predictions to untrained new data. As the complexity of the neural network increases, this overfitting phenomenon also becomes more severe. In this paper, by taking statistical viewpoint, we proposed an integrative process for neural network design and model selection method in order to improve generalization performance. At first, by using the natural gradient learning with adaptive regularization, we try to obtain optimal parameters that are not overfilled to training data with fast convergence. By adopting the natural pruning to the obtained optimal parameters, we generate several candidates of network model with different sizes. Finally, we select an optimal model among candidate models based on the Bayesian Information Criteria. Through the computer simulation on benchmark problems, we confirm the generalization and structure optimization performance of the proposed integrative process of teaming and model selection.

• KIPS Transactions on Software and Data Engineering
• /
• v.8 no.12
• /
• pp.491-498
• /
• 2019

Digital twin is a technology that virtualizes physical objects of the real world on a computer. It is used by collecting sensor data through IoT, and using the collected data to connect physical objects and virtual objects in both directions. It has an advantage of minimizing risk by tuning an operation of virtual model through simulation and responding to varying environment by exploiting experiments in advance. Recently, artificial intelligence and machine learning technologies have been attracting attention, so that tendency to virtualize a behavior of physical objects, observe virtual models, and apply various scenarios is increasing. In particular, recognition of each robot's motion is needed to build digital twin for co-robot which is a heart of industry 4.0 factory automation. Compared with modeling based research for recognizing motion of co-robot, there are few attempts to predict motion based on sensor data. Therefore, in this paper, an experimental environment for collecting current and inertia data in co-robot to detect the motion of the robot is built, and a motion prediction model based on the collected sensor data is proposed. The proposed method classifies the co-robot's motion commands into 9 types based on joint position and uses current and inertial sensor values to predict them by accumulated learning. The data used for accumulating learning is the sensor values that are collected when the co-robot operates with margin in input parameters of the motion commands. Through this, the model is constructed to predict not only the nine movements along the same path but also the movements along the similar path. As a result of learning using SVM, the accuracy, precision, and recall factors of the model were evaluated as 97% on average.