• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.35 no.2
• /
• pp.125-130
• /
• 2011

Since conventional optimization that is classified as a deterministic method does not consider the uncertainty involved in a modeling or manufacturing process, an optimum design is often determined to be on the boundaries of the feasible region of constraints. Reliability-based design optimization is a method for obtaining a solution by minimizing the objective function while satisfying the reliability constraints. This method includes an optimization process and a reliability analysis that facilitates the quantization of the uncertainties related to design variables. Moment-based reliability analysis is a method for calculating the reliability of a system on the basis of statistical moments. In general, on the basis of these statistical moments, the Pearson system estimates seven types of distributions and determines the reliability of the system. However, it is technically difficult to practically consider the Pearson Type IV distribution. In this study, we propose an enhanced Pearson Type IV distribution based on a kriging model and validate the accuracy of the enhanced Pearson Type IV distribution by comparing it with a Monte Carlo simulation. Finally, reliability-based design optimization is performed for a system with type IV distribution by using the proposed method.

• Journal of IKEEE
• /
• v.19 no.4
• /
• pp.491-499
• /
• 2015

Silicon Carbide(SiC) has large advantage in high temperature and high voltage applications because of its high thermal conductivity and large band gap energy. When using SiC to design power semiconductor devices, edge termination techniques have to be adjusted for its maximum breakdown voltage characteristics. Many edge termination techniques have been proposed, and the most appropriate technique for SiC device is Junction Termination Extension(JTE). In this paper, the change of breakdown voltage efficiency ratio according to the change of doping concentration and passivation oxide charge of each JTE techniques is demonstrated. As a result, the maximum breakdown voltage ratio of Single Zone JTE(SZ-JTE), Double Zone JTE(DZ-JTE), Multiple Floating Zone JTE(MFZ-JTE), and Space Modulated JTE(SM-JTE) is 98.24%, 99.02%, 98.98%, 99.22% each. MFZ-JTE has the smallest and SZ-JTE has the largest sensitivity of breakdown voltage ratios according to the change of JTE doping concentration. Additionally the degradation of breakdown voltage due to the passivation oxide charge is analyzed, and the sensitivity is largest in SZ-JTE and smallest in MFZ-JTE, too. In this paper, DZ-JTE and SM-JTE is the best efficiency JTE techniques than MFZ-JTE which needs large doping concentration in short JTE width.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.15 no.3
• /
• pp.13-22
• /
• 2012

The spatial structure analysis using the existing space syntax has a weakness only to apply the road factor. Hence, it is necessary to perform a new spatial structure analysis considering the regional characteristics factor as well as the road factor. The purpose of this study is to propose a modified spatial structure analysis which adopts road and regional characteristics factors using the existing space syntax and GIS and then to apply it to the spatial structure analysis of the urbanized areas in Hapdeok town. For the new analysis, the experts' survey has been conducted to derive the spatial structure factors of the area and their relative weights. Specifically, it drew the road factor as well as the regional characteristics factors such as road, district, land value, and building density. This study shows that the existing spatial structure analysis considering only the road factor does not reflect the actual spatial structure but the modified spatial structure analysis considering the road factor and the regional characteristics factors more accurately analyzes the spatial structure of Hapdeok town. Hapdeok is not an area where road networks were developed along main development axis and its centrality was changed, but an area where road networks were developed by effectiveness and equity and its centrality was changed by downtown decline. The towns like Hapdeok is more suitable for applying the modified spatial structure analysis method considering the regional characteristics factors. Further research is required for selecting characteristics factors and appropriate parameters in the modified spatial structure analysis. Further case studies await in other cities for conducting its verification.

• Proceedings of the Computational Structural Engineering Institute Conference
• /
• 2010.04a
• /
• pp.618-619
• /
• 2010

LED는 기존의 발광원에 비해 훨씬 높은 파워와 효율성으로 인해 최근 들어 각종 조명이나 교통신호 등에서 사용이 급증하고 있다. LED 재료를 위해 지금까지 여러가지가 연구되어 왔는데, 갈륨 질화물 (Gallium Nitride, GaN)에 기반한 시스템이 최근들어 가장 큰 관심을 받고 있다. GaN 방식은 열적으로 매우 안정성이 있고, 1.9 ~ 6.2 eV 범위의 넓은 밴드의 Gap, 그리고 인듐이나 알루미늄과 결합하여 청, 녹, 백색등의 다양한 빛을 발생할 수 있는 장점을 가지고 있다. 예를 들어 청색 LED는 광학 방식의 기록매체에, 백색 LED는 기존의 조명램프의 대체용으로 활용이 가능하다. 이러한 장점 덕분에 GaN기반 LED 시장은 1994년에 최초로 상용화 된 이래 최근 급격한 성장을 보여 왔다. 그러나 GaN은 다른 III~V 타입의 반도체 재료와는 달리 재료가 성장하기 위해 사파이어와 같은 별도의 기판을 필요로 하는 문제가 있다. 이것은 결국 전위발생과 같은 격자의 부조화 같은 문제를 야기하여 결국 LED의 성능을 떨어뜨리는 요인이 된다. 이러한 문제를 해결하기 위해 HVPE(Hydride Vapor Phase Epitaxy) 방법이 개발되었는데, 이 방법은 시간당 100 미크론의 매우 빠른 성장속도로 높은 두께의 레이어를 만드는 장점이 있다. 이렇게 성장된 GaN 레이어는 베이스 기판에서 쉽게 분리되어 활용이 가능하다. 그러나 HVPE 기술은 성장 공정에서 두께를 균일하게 만들도록 제어하는 것이 매우 어렵다는 문제가 있다. 따라서 HVPE 방식에서는 이러한 조건을 만족시키기 위해 반응현상에 대한 물리적 해석을 토대로 공정조건을 정밀하게 설계해야 한다. 이를 위해 최근에 실험 또는 시뮬레이션을 활용하여 이러한 공정조건을 향상시키기 위한 여러 연구가 진행되었다. 본 연구에서는 이러한 연구의 일환으로 반응로에 투입되는 여러 기체의 유량과 존별 주변온도 조건을 입력변수로 하고, 이들이 GaN 성장에 미치는 영향을 분석하였다. HVPE 시스템에서 가장 이상적인 목표는 반응기체가 층류유동을 유지하면서 대부분의 반응이 기판위에서 이뤄지며, 기판위에서 성장되는 재료의 두께가 균일하게 되는 것이다. 입력변수들이 이러한 결과에 어떠한 영향을 미치는 지 분석하기 위해 전산유체역학(CFD, Computational Fluid Dynamics)을 수행하는 상용코드 FLUENT를 사용하였다. 보다 실제에 가까운 해석을 위해서는 기체간의 화학반응을 포함해야 하나, 해석의 편의와 효율을 위해 본 연구에서는 열 및 유동해석만을 수행하였다. 한편 실제 반응로의 우수성은 성장속도와 두께분포의 균일도를 통해 평가된다. CFD 해석을 통해 이들을 분석하기 위해 기존에 수행한 실험조건을 해석하고 해석결과의 유동패턴/압력분포를 실험결과의 성장속도/두께분포와 비교하고, 이중에서 관련성이 높은 해석결과변수를 우수성 평가에 활용하였다. 기존의 실험결과를 토대로 이러한 중요 결과변수와 함께 이들에 대한 목표값이 도출되고 나면, 입력 공정조건 - 사용기체의 유량과 주변온도 조건 - 에 대해 실험계획(DOE,Design of Experiment)을 수립하고 목표성능을 구현하기 위한 최적설계를 수행할 수 있다. 일반적으로 CFD를 통해 최적의 설계나 공정조건을 탐색하는 작업은 1회의 CFD 계산시간이 매우 오래 소요되기 때문에 쉽지 않다. 그러나 본 연구에서는 CFD와 DOE의 적절한 조합을 통해 적은 수의 해석을 가지고도 원하는 결과를 효율적으로 얻는 것이 가능함을 입증하고자 한다. 본 발표에서는 아직 이러한 연구가 완성되지 않은 시점에서 제반 연구개요를 소개하고 현 시점까지의 연구 결과 및 향후 계획을 소개하고자 한다.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.27 no.4
• /
• pp.281-288
• /
• 2014

As a gas explosion is the most fatal accident in shipbuilding and offshore plant industries, all safety critical elements on the topside of offshore platforms should retain their integrity against blast pressure. Even though many efforts have been devoted to develop blast-resistant design methods in the offshore engineering field, there still remain several issues needed to be carefully investigated. From a procedure for calculation of explosion design pressure, impulse of a design pressure model having completely positive side only is determined by the absolute area of each obtained transient pressure response through the CFD analysis. The negative pressure phase in a general gas explosion, however, is often quite considerable unlike gaseous detonation or TNT explosion. The main objective of this study is to thoroughly examine the effect of the negative pressure phase on structural behavior. A blast wall for specific FPSO topside is selected to analyze structural response under the blast pressure. Because the blast wall is considered an essential structure for blast-resistant design. Pressure time history data were obtained by explosion simulations using FLACS, and the nonlinear transient finite element analyses were performed using LS-DYNA.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.37 no.7B
• /
• pp.489-498
• /
• 2012

Recently, as the interest in Green IT is exponentially increased, EPON with sleep mode has been studied to reduce energy consumption in access network. In oder to guarantee cyclic sleep for ONU(Optical Network Units), EPON with sleep mode transmits upstream and downstream data at the same time. However, since conventional algorithms for sleep mode in EPON allocate bandwidth to each ONU according to upstream bandwidth request, the QoS of downstream data is not guaranteed when the offered load of OLT is larger than that of ONU. In this paper, we propose a bandwidth allocation algorithm for improving QoS in EPON with sleep mode. The proposed algorithm compares its size with an upstream request of ONU when a downstream buffer in the OLT exceeds a QoS threshold. And then it allocates selectively a bandwidth that satisfies the required QoS between the bandwidth request of ONU and OLT. Therefore, the proposed algorithm can save energy through cyclic sleep of ONUs while guaranteeing the QoS of up/downstream data. In order to evaluate the proposed algorithm, we perform simulation in terms of total sleep time of ONUs, queueing delay between OLT and ONU, and the utilization of allocated bandwidth at OLT through OPNET.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.36 no.3A
• /
• pp.232-239
• /
• 2011

IEEE 802.11n standard provides a framework for new link adaptation. A station can request that another station provide a Modulation and Coding Scheme (MCS) feedback, to fully exploit channel variations on a link. However, if the time elapsed between MCS feedback request and the data frame transmission using the MCS feedback becomes bigger, the previously received feedback information may be obsolete. In that case, the effectiveness of the feedback-based link adaptation is compromised. If a station can estimate how fast the channel quality to the target station changes, it can improve accuracy of the link adaptation. The contribution of this paper is twofold. First, through a thorough NS-2 simulation, we show how the coherence time affects the performance of the MCS feedback based link adaptation of 802.11n networks. Second, this paper proposes an effective algorithm for coherence time estimation. Using Allan variance information statistic, a station estimates the coherence time of the receiving link. A proposed link adaptation scheme considering the coherence time can provide better performance.

• Journal of Biomedical Engineering Research
• /
• v.25 no.6
• /
• pp.489-496
• /
• 2004

Insertion of cannulae into vessels may disturb the blood flow doing non-physiological load and stress on blood cells such that ADP may increase and result in hemolysis. Authors used the computational method to simulate the 3-dimensional blood flow inside of the cannula using numerical method. We limited the research to within the drainage cannulae with side holes inserted through the human vein. In this paper, 9 different cannulae with side holes categorized by the number of side holes of 4, 12, and 20, and also categorized by the array type of side holes of staggered array, in-line array, and alternative in-line array were studied and compared to the cannula with no side holes by using CFD analysis. We evaluated the flow rate, the wall shear stress, and the shear rate and compared them with one another to estimate the effect of the side holes. The flow rate is not proportional to the number of the side holes. However, larger number of side holes can reduce the mean shear rate. Both the number and the array type of side holes play an important role on the fluid dynamics of the blood flow in cannulae.

• Journal of KIISE:Information Networking
• /
• v.29 no.1
• /
• pp.65-76
• /
• 2002

QoS routing is a routing technique for finding feasible path that satisfies QoS requirements required by application programs. Since QoS routing determines such paths in terms of dynamic network state, it satisfies the requirement of applications and increases the utilization of the network. The overhead is, however, generated by routers to exchange the information of the dynamic state of network. In order to reduce this protocol overhead, a timer based update mechanism is proposed in which router checks the change of the network status periodically and network state information is exchanged if the change is greater than a certain value. Using large update period makes, though, routing performance irresponsive to the parameters which determine the update of the network state of the router. In addition to this, large update period may result in inaccurate network state information at routers and cause resource reservation failure. The resource reservation failure generates additional overhead to cancel the resource reservation of the part of the path. In this paper, we propose mechanisms enhancing the existing network state update policy with respect to these two problems. Performance of the proposed schemes are evaluated through a course of simulation.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.41 no.2
• /
• pp.263-268
• /
• 2016

A rapid increase of data amount, used in ship-to-ship transmission of safety and logistics information, ships in the inland sea have trouble transmitting real-time information transmission due to an increase in traffic load caused by data transmitted by land station and offshore ships. In this study, therefore, communication is carried out by adaptively controlling the detailed beam width based on the distribution of offshore ships in land station durable in marine environment. Then after the adaptive sector system enabling real-time communication support between ships concentrated in an inland sea and land station is applied, the performance verification is conducted based on the respective Call Blocking Rates of an omnidirectional antenna, fisted sector system, and adaptive sector system. The performance verification result shows that adaptive sector system has better performance than the fixed sector system as the density of ship, q value, increases, and that the smaller the beam width is, the better performance of adaptive sector system will be.