• Proceeding of EDISON Challenge
• /
• 2016.03a
• /
• pp.116-119
• /
• 2016

Chlorosulfolipids(CSLs)는 1960 년대에 해조류에서 발견되었다. 하지만 당시 기술력으로는 해조류에서 CSLs 를 추출해 내는 것이 불가능하여 연구가 중단 되었다. 그 후로 40 년 뒤 2009 년이 되어서야 CSLs 를 추출할 수 있게 되었다. CSLs 가 독성을 지니고 있다는 연구가 보고 되어 왔다. 하지만 이 CSL 가 형성하는 구조를 실험으로 알수가 없고 아직까지 학회에 보고 된 적이 없다. 따라서 본 연구원이 MD 시뮬레이션을 이용하여 CSLs 가 형성하는 세포막의 구조를 알아보기 위하여 Coarse-grain 모델을 이용한 CSLs 의 Self-Assembly 연구와 이 결과로 인해 얻은 정보로 atomistic 모델을 만들어서 MD 시뮬레이션을 수행하였다.

• Proceeding of EDISON Challenge
• /
• 2016.03a
• /
• pp.404-410
• /
• 2016

Compared to wheel locomotion, walking has many advantages : Better to cross over obstacles, the contact with ground is in a determined point, the ground is damaged less. Because Theo Jansen mechanism can make walking motion that is very soft, there are many researches about that mechanism. In this paper, We designed walking robot based on Theo Jansen mechanism. Most important design factor is velocity and stability. First of all, we considered kinematics knowledge and then, we made a new model by using simulation. Finally we developed the model by solving few design problems.

• Proceeding of EDISON Challenge
• /
• 2016.03a
• /
• pp.75-79
• /
• 2016

A helical nanosturctrue can be obtained by self-assembly method. Utilizing DPD simulation coarse-grained model, we patterned 2D layer nanosheets with repeated diagonal defects and grafts, and programed to self-roll into hollow helix structure. The defected pattern side caused anisotropy, and formed helix or helix-like structure. This opens the possibility to control the helix pitch or cavity radius. In this work, we designed several patterns about diagonal defect with a variety of defect side densities and defect widths and then simulation was carried out. Thus, our results have that parameters are affecting self-assembly of nanosheets and their conformation.

• Proceeding of EDISON Challenge
• /
• 2016.03a
• /
• pp.560-565
• /
• 2016

최근 대형 컨테이너선의 개발이 지속적으로 이루어짐에 따라 슬래밍에 의한 선수 및 선미의 구조안정성 문제가 대두되고 있지만 설계 단계에서 슬래밍에 대해 고려하기에는 현상의 복잡성으로 인해 어려움이 많았다. 이를 위해 KRISO에서 시행된 WILS JIP의 선수 단면 형상 및 선미 단면인 쐐기 형상으로 격자를 생성하여 EDISON CFD 다상유동 해석자를 통해 수치해석을 시도하였다. 기존 방식과 달리 계산 시간 절감을 위하여 격자 변형 기법을 적용하지 않고 모형 시험결과를 기반으로 한 유입류 조건을 설정하여 입수 충격 문제를 해석해보았다. 그 결과, 선미 형상의 경우 선행연구와 유사하게 실험 결과에 근접한 유체 충격력을 정량적으로 얻어낼 수 있었다. 선수 형상의 경우에서는 구상 선수로 인해 파생되는 센서 위치별 충격력의 변화를 확인할 수 있었으며, 실제 유동에 가까운 유동 형상과 슬래밍에 의한 충격력을 개략적으로 구할 수 있었다.

• Proceeding of EDISON Challenge
• /
• 2016.03a
• /
• pp.155-162
• /
• 2016

This study attempts to reveal structures of two-dimensional ring polymer solutions in various polymer concentrations ranging from dilute to concentrated regime. Polymer sizes, single molecule structure factors, bond correlation functions and monomer density distribution functions from center of mass are given in order to clarify the polymer structures. Our study shows that a ring in dilute solution maintain pseudo-circular structure with self-avoiding walk (SAW) statistics, and it seems to be composed of two connecting SAW linear chains. In semidilute solutions, ring polymers are not entangled with each other and adopt collapsed configurations. Such assumption of collapsed structures in the semidilute regime gives an overlap concentration of ${\varphi}^*{\sim}N^{-1/2}$ where N is degree of polymerization. By normalizing the polymer concentration by these overlap concentration, we find universal behaviors of polymer sizes and structure factors regardless of N.

• International Journal of Aeronautical and Space Sciences
• /
• v.13 no.1
• /
• pp.1-13
• /
• 2012

Over 3.8 billion years, through evolution nature came up with many effective continually improving solutions to its challenges. Humans have always been inspired by nature capabilities in problems solving and innovation. These efforts have been intensified in recent years where systematic studies are being made towards better understanding and applying more sophisticated capabilities in this field that is increasingly being titled biomimetics. The ultimate challenge to this field is the development of humanlike robots that talk, interpret speech, walk, as well as make eye-contact and facial expressions with some capabilities that are exceeding the original model from nature. This includes flight where there is no creature that is as large, can fly as high, carry so heavy weight, fly so fast, and able to operate in extreme conditions as the aircraft and other aerospace systems. However, there are many capabilities of biological systems that are not feasible to mimic using the available technology. In this paper, the state-of-the-art of some of the developed biomimetic capabilities, potentials and challenges will be reviewed.

• Proceedings of the KSRS Conference
• /
• 2003.11a
• /
• pp.57-59
• /
• 2003

Digital surface model generation from IKONOS stereo imagery is a new challenge in photogrammetric community, especially when the satellite company does not provide the raw data as well as their ancillary ephemeris data. In this paper we utilized an estimated relief displacement azimuth and the nominal collection elevation data included in the metadata file to correct the relief displacement of GCPs, together with a linear transformation for geometric modeling of IKONOS imagery. Space intersection is performed by the trigonometric intersection assuming a parallel projection of IKONOS imagery due to its small FOV and frame size. In the experiment, less than 2-meters of RMSE in orbit modeling is achieved denoting the potential positioning accuracy of the IKONOS stereo imagery.

• International Journal of Ocean System Engineering
• /
• v.2 no.2
• /
• pp.97-105
• /
• 2012

A all floating structures operating within a limited area require, stationkeeping to maintain the motions of the floating structure within permissible limits. In this study, methods for selecting and optimizing the mooring system Caisson for floating wind turbines in shallow water are investigated. The design of the mooring system is checked against the governing rules and standards. Adequately verifying the design of floating structures requires both numerical simulations and model testing, the combination of which is referred to as the hybrid method of design verification. The challenge in directly scaling moorings for model tests is the depth and spatial limitations of wave basins. It is therefore important to design and build equivalent mooring systems to ensure accurate static properties (global restoring forces and global stiffness).

• International Journal of Computer Science & Network Security
• /
• v.21 no.8
• /
• pp.212-218
• /
• 2021

The Proportional Integral Derivative (PID) controller is the most popular industrial controller and more than 90% process industries use this controller. During the past 50 years, numerous good tuning methods have been proposed for Single Input Single Output Systems. However, design of PI/PID controllers for multivariable processes is a challenge for the researchers. A comparative study of three PID controllers design methods has been carried-out. These methods include the DS (Direct Synthesis) method, IMC (Internal model Control) method and ETF (Effective Transfer Function) method. MIMO PID controllers are designed for a number of 2×2, 3×3 and 4×4 process models with multiple delays. The performance of the three methods has been evaluated through simulation studies in Matlab/Simulink environment. After extensive simulation studies, it is found that the Effective Transfer Function (ETF) Method produces better output responses among two methods. In this work, only decentralized methods of PID controllers have been studied and investigated.

• Journal of Sensor Science and Technology
• /
• v.30 no.5
• /
• pp.290-294
• /
• 2021

Experts have designed popular and successful model architectures, which, however, were not the optimal option for different scenarios. Despite the remarkable performances achieved by deep neural networks, manually designed networks for classification tasks are the backbone of object detection. One major challenge is the ImageNet pre-training of the search space representation; moreover, the searched network incurs huge computational cost. Therefore, to overcome the obstacle of the pre-training process, we introduce a network adaptation technique using a pre-trained backbone model tested on ImageNet. The adaptation method can efficiently adapt the manually designed network on ImageNet to the new object-detection task. Neural architecture search (NAS) is adopted to adapt the architecture of the network. The adaptation is conducted on the MobileNetV2 network. The proposed NAS is tested using SSDLite detector. The results demonstrate increased performance compared to existing network architecture in terms of search cost, total number of adder arithmetics (Madds), and mean Average Precision(mAP). The total computational cost of the proposed NAS is much less than that of the State Of The Art (SOTA) NAS method.