• Korean Journal of Cognitive Science
• /
• v.10 no.4
• /
• pp.63-70
• /
• 1999

Reconstruction of collision accidents is to analyze the cause of accidents and collision behavior using available information from vehicle accident circumstances. This paper introduces a collision reconstruction system which is developed to be applicable to traffic accident reconstruction. Our System combines both quantitative and qualitative collision models so as to compensate for weaknesses in each with strengths of each other. I It provides accurate predictions and causal explanations of the collision behavior. During r reverse analysis of collision. qualitative simulation is used to verify a hypothesis and to detect any conflict in early stage of reconstruction. It is implemented and applied to real car-to-car collision accidents. The test results verify the reliabilities of our techniques.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.18 no.2
• /
• pp.23-28
• /
• 2004

In EIT, various image reconstruction algorithms have been used in order to compute the internal resistivity distribution of the unknown object with its electric potential data at the boundary. Mathematically the EIT image reconstruction algorithm is a nonlinear ill-posed inverse problem. In this paper, a SPSA approach is proposed for the solution of the EIT image reconstruction. Results of numerical experiments of EIT solved by the SPSA approach are presented and compared to that obtained by the modified Newton-Raphson(mNR) method.

• Korean Journal of Optics and Photonics
• /
• v.7 no.4
• /
• pp.370-374
• /
• 1996

A Superstructure Grating(SSG) Distributed-Bragg-Reflector(DBR) laser has a broad tuning range with a good mode suppression ratio. However, gaps of channel are observed in the wavelength-tuning characteristics of an SSGDBR laser which employs linearly-chirped DBR mirrors. We found by numerical simulation that the gaps may be attributed to the nonuniform reflection-peak heights of a linearly-chirped DBR mirrors. We propose a nonlinearly chirped grating DBR mirror structure that makes reflection-peak heights almost uniform. Therefore a nonlinearly chirped grating structure can be employed in an extended tuning range semiconductor laser to achieve gap-free tuning and low threshold current operation simultaneously.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.31 no.4
• /
• pp.320-329
• /
• 2007

Electrokinesis and dielectrophoresis are important transport phenomena produced by external electric field applied to a microchannel containing a conductive fluid. We developed a CFD code to predict electrokinetic and dielectrophoretic flows in a microchannel with a uniform circular post array. Using the code, we calculated particle velocities driven by electrokinesis and dielectrophoresis, and conducted Monte Carlo simulations to visualize the particle motions. The code was validated by comparing the results with those from previous studies in literature. At a low electric field, electrokinesis and diffusion is the dominant transport mechanism. At a moderate electric field, dielectrophoresis is balanced with electrokinesis and diffusion, resulting in flowing filaments of particles in the microchannels. However, dielectrophoresis overwhelms the flow at a high electric field and traps particles locally. These results provide useful insight for optimizing design parameters of a microfluidic chip for biochemical analysis, especially for development of on-chip sample pretreatment techniques using electrokinetic and dielectrophoretic effects.

• The KIPS Transactions:PartB
• /
• v.8B no.3
• /
• pp.231-236
• /
• 2001

일반적인 다층 신경망에서 가중치의 갱신 알고리즘으로 사용하는 오류 역전과 방식은 가중치 갱신 결과를 고정된(fixed) 한 개의 값으로 결정한다. 이는 여러 갱신의 가능성을 오직 한 개의 값으로 고정하기 때문에 다양한 가능성들을 모두 수용하지 못하는 면이 있다. 하지만 모든 가능성을 확률적 분포로 표현하는 갱신 알고리즘을 도입하면 이런 문제는 해결된다. 이러한 알고리즘을 사용한 베이지안 신경망 모형(Bayesian Neural Networks Models)은 주어진 입력값(Input)에 대해 블랙 박스(Black-Box)와같은 신경망 구조의 각 층(Layer)을 거친 출력값(Out put)을 계산한다. 이 때 주어진 입력 데이터에 대한 결과의 예측값은 사후분포(posterior distribution)의 기댓값(mean)에 의해 계산할 수 있다. 주어진 사전분포(prior distribution)와 학습데이터에 의한 우도함수(likelihood functions)에 의해 계산한 사후확률의 함수는 매우 복잡한 구조를 가짐으로 기댓값의 적분계산에 대한 어려움이 발생한다. 따라서 수치해석적인 방법보다는 확률적 추정에 의한 근사 방법인 몬테 칼로 시뮬레이션을 이용할 수 있다. 이러한 방법으로서 Hybrid Monte Carlo 알고리즘은 좋은 결과를 제공하여준다(Neal 1996). 본 논문에서는 Hybrid Monte Carlo 알고리즘을 적용한 신경망이 기존의 CHAID, CART 그리고 QUEST와 같은 여러 가지 분류 알고리즘에 비해서 우수한 결과를 제공하는 것을 나타내고 있다.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.37 no.8A
• /
• pp.618-631
• /
• 2012

In this paper polarization adaptivity on transmit has been used to enhance the received signals directed to a pre-selected receiver in a near-field multi-input multi-output (MIMO) environment. The objective here is to select a set of weights on the transmitting antennas adapted to individual receivers based on the principles of reciprocity. Using the polarization properties, when the number of receiving antennas is greater than the number of transmitting antennas, the transmitted signal may be directed more to a particular receiver location while simultaneously minimizing the reception signal strength at other receivers. Numerical simulations have been made to illustrate the novelty of the proposed approach.

• Explosives and Blasting
• /
• v.26 no.1
• /
• pp.1-5
• /
• 2008

In this study, the structural behavior accorded by pre-weakening before demolishing a cylindrical structure was investigated to ensure structural stability using 3 dimensional applied element method (3D AEM). An opening angle of crushed area for the pre-weakening was changed to examine the displacement behavior of the cylindrical structure. This study shows that the crushing range for pre-weakness must be below 40% to void the collapse of the cylindrical silo structure.

• Korean Journal of Optics and Photonics
• /
• v.11 no.3
• /
• pp.171-178
• /
• 2000

We present a numerical model which enables to analyze $1.47mu$m amplification band characteristics for thulium doped fluoride fiber amplifiers. We focused on upconversion pumping scheme, thus many transitions affecting $1.47mu$m band amplification was considered simultaneously. Backward propagating waves and transverse mode were also considered in the model. The parameters for modeling were then acquired using published experimental data and related theories such as Judd-Ofelt theory for radiative transition, empirical energy gap law for non-radiative transition, and McCumber relations for cross-sections. The simulation showed well-matched results with experiment and internal dynamics. amics.

• Journal of Korean Society of Steel Construction
• /
• v.10 no.2 s.35
• /
• pp.279-290
• /
• 1998

An existing Damage Index Method is verified to demonstrate its feasibility for detecting structural damage in truss bridges (1) for which modal parameters are available for a few modes of vibration and (2) for which baseline modal information is not available from its as-built state. The theory of approach to detect locations of damage and to identify baseline modal model is summarized on the basis of system identification theory and modal sensitivity theory. The feasibility of the Damage Index Method is demonstrated using a numerical example of a truss bridge with 11 subsystems of 211 members and for which only two modes of vibration were recorded for post-damaged state.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.33 no.9
• /
• pp.81-88
• /
• 2005

This paper analyzes the fuel consumption for the momentum dumping of the COMS which has a single solar array system. First, numerical analyses are conducted to find an optimal momentum dumping time considering the COMS configuration. It is assumed that the momentum dumping is conducted once a day and at a fixed time of a day. Secondly, in an effort to reduce the momentum dumping fuel consumption, this paper proposes a new approach which combines the momentum dumping and the ordinary north/south stationkeeping. Finally, to evaluate the proposed technique, the stationkeeping simulations are conducted and analyzed.