• 제어로봇시스템학회논문지
• /
• 제5권2호
• /
• pp.189-199
• /
• 1999

This paper presents an optimization algorithm for a stable Self Dynamic Neural Network(SDNN) using genetic algorithm. Optimized SDNN is applied to a problem of controlling nonlinear dynamical systems. SDNN is dynamic mapping and is better suited for dynamical systems than static forward neural network. The real-time implementation is very important, and thus the neuro controller also needs to be designed such that it converges with a relatively small number of training cycles. SDW has considerably fewer weights than DNN. Since there is no interlink among the hidden layer. The object of proposed algorithm is that the number of self dynamic neuron node and the gradient of activation functions are simultaneously optimized by genetic algorithms. To guarantee convergence, an analytic method based on the Lyapunov function is used to find a stable learning for the SDNN. The ability and effectiveness of identifying and controlling a nonlinear dynamic system using the proposed optimized SDNN considering stability is demonstrated by case studies.

• Bulletin of the Korean Chemical Society
• /
• 제20권11호
• /
• pp.1323-1328
• /
• 1999

Interdiffusion between two partially miscible polymers of similar chemical structures with different molecular weights is characterized theoretically by using the reptation model for the interdiffusion. This model provides more reliable results than the early Rouse model for same molecular weights, compared with the experiments. Furthermore, by introducing the molecular weight ratio R into the reptation model, we can see the dynamic effect of molecular weight on the diffusion behaviors of the asymmetric system. Near the critical point the diffusion behaviors of asymmetric binary polymer mixtures are well characterized by the interfacial width W(t), the mass transport M(t) for the different values of the Flory Χ parameter and different molecular weight ratios ofpolymers of the diffusion couple. These two quantities and composition profiles by this model give betteragreement with experiments.

• 전자공학회논문지C
• /
• 제36C권5호
• /
• pp.76-85
• /
• 1999

신경회로망을 동적 정보처리에 응용하기 위해서는 비대칭 결합 신경회로망에서 생성되는 동적 상태천이에 관한 직관적 이해가 필요하다. 본 논문에서는 각 뉴런이 최근접 뉴런에만 이진화된 결합하중 +1 및 -1로 연결된 순환결합형 신경회로망의 동적인 상태천이 특성을 해석하였다. 상태천이 해석 알고리즘을 이용한 시뮬레이션 결과, 네트워크는 고정점, 베이슨을 갖는 리미트사이클 및 베이슨이 없는 리미트사이클의 3가지 어트랙터를 가진다. 또한, 네트워크에서 생성되는 리미트사이클의 수와 주기를 이론적으로 해석하여 정식화하고, 리미트사이클을 구성하는 상태벡터의 필요조건을 나타내었다. 이론 해석의 결과는 네트워크에서 생성되는 리미트사이클의 수가 뉴런(소자)의 수 n에 대해서 지수 함수적으로 증가함을 보여준다. 따라서 순환 결합형 신경회로망은 많은 동적 정보를 저장할 수 있는 메모리 시스템으로 이용할 수 있다.

• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송∙미디어공학회 2017년도 하계학술대회
• /
• pp.83-84
• /
• 2017

We propose a novel high dynamic range (HDR) video synthesis algorithm from alternatively exposed low dynamic range (LDR) videos. We first estimate correspondences between input fames using an illumination invariant descriptor. Then, we synthesize an HDR frame with the weights computed to maximize detail preservation in the output HDR frame. Experimental results demonstrate that the proposed algorithm provides high-quality HDR videos without noticeable artifacts.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2000년도 제15차 학술회의논문집
• /
• pp.384-384
• /
• 2000

In adaptive neuro-control, neural networks are used to approximate the unknown plant nonlinearities. Until now, most of the papers in the field of controller design fur nonlinear system using neural networks considers the affine system with fixed number of neurons. This paper considers nonaffne nonlinear systems and dynamic variation of the number of neurons. Control laws and adaptive laws for weights are established so that the whole system is stable in the sense of Lyapunov.

• 한국조사연구학회지:조사연구
• /
• 제12권3호
• /
• pp.173-186
• /
• 2011

본 연구에서는 2003년 시작된 한국청소년패널조사 중학교 2학년 패널자료 분석을 위해 필수적인 각 연도별 횡단면, 종단면 가중치 산출방안을 소개하고 있다. 패널 모집 당시 추출된 표본이 대표하는 모집단의 종단면적 변화 분석을 위한 종단면 가중치는 로지스틱 회귀분석을 이용한 무응답 보정과 사후 층화를 통해 산출되었으며, 조사 연도의 표본 대응 모집단 분석을 위해 필수적인 횡단면 가중치는 전년도 대비 응답률과 사후 층화를 통해 산출되었다.

• 한국전산구조공학회:학술대회논문집
• /
• 한국전산구조공학회 1997년도 가을 학술발표회 논문집
• /
• pp.8-14
• /
• 1997

The dynamic responses of highway bridges are varying depending on the features of either traveling vehicles or bridges. In this study, the probabilistic characteristics of dynamic amplification factors of highway bridges due to traveling heavy vehicles have been examined through analytical simulation processes. The truck with tandem axle and tractor with semitrailer are selected as the representative heavy vehicles, which are modeled with three dimensional 7-DOF and 12-DOF models, respectively. The analytical results have been compared with the experimental results of dynamic loading tests and the validity of the analytical models has been examined. Parametric studies on the means and extreme values of amplification factors have been performed with various traffic conditions such as vehicle types, vehicle weights, surface profiles, vehicle velocity, etc.

• 한국도로학회논문집
• /
• 제16권1호
• /
• pp.21-30
• /
• 2014

PURPOSES : Evaluation of input parameters determination procedure for dynamic analysis of aggregates in DEM. METHODS : In this research, the aggregate slump test and angularity test were performed as fundamental laboratory tests to determine input parameters of spherical particles in DEM. The heights spreads, weights of the simple tests were measured and used to calibrate rolling and static friction coefficients of particles. RESULTS : The DEM simulations with calibrated parameters showed good agreement with the laboratory test results for given dynamic condition. CONCLUSIONS : It is concluded that the employed calibration method can be applicable to determine rolling friction coefficient of DEM simulation for given dynamic conditions. However, further research is necessary to connect the result to the behavior of aggregate in packing and mixing process and to refine static friction coefficient.

• Multiscale and Multiphysics Mechanics
• /
• 제1권4호
• /
• pp.327-340
• /
• 2016

Poly(${\epsilon}$-caprolactone) (PCL) diol, with good biodegradation and biocompatibility, is one of the widely used soft segments (SSs) in composing bio-polyester-urethanes (Bio-PUs), which show great potential in both biomedical and tissue engineering applications. Properties of Bio-PUs are tunable by combining SS monomers with different molecular weights, structures, modifications, and ratio of components. Although numbers of research have reported many Bio-PUs properties, few studies have been done at the molecular scale. In this study, we use molecular dynamic (MD) simulation to construct atomistic models for two commonly used PCL diol SSs with different molecular weights 1247.58 Da and 1932.42 Da. We compare the simulation results by using two widely used classical force fields for organic molecules: Consistent Valence Force Field (CVFF) and CHARMM General Force Field (CGenFF), and discuss the validity and accuracy. Melt density, volume, polymer conformations, transition temperature, and mechanical properties of PCL diols are calculated and compared with experiments. Our results show that both force fields provide accurate predictions on the properties of PCL diol system at the molecular scale and could help the design of future Bio-PUs.

• 한국염색가공학회지
• /
• 제11권4호
• /
• pp.1-7
• /
• 1999

Urethane-acrylate propelymers for secondary coating of optical fiber and high - performance material were prepared from the 4,4'-diphenylmethane diisocyanate(MDI), poly(tetramethylene oxide)glycol(PTMG, Mw 650 or 1000), 1,6-hexanediol(HD), 2-hydroxyethyl acrylate(HEA), and dibutyltin dilaurate as a catalyst. UV-Curable polyurethane acrylates were formulated from the urethane-acrylate prepolymers, three types of reactive diluents(DTs) having mono-, di-, and trifunctional-phenoxyethyl acrylate(PEA), hexanediol diacrylate(HDDA), and trimethylolpropane triacrylate(TMPTA), and 1-hydtoxycyclohexyl phenyl ketone(Irgacure 184) as a photoinitiator. The UV-cured films of polyurethane acrylates were obtained by curing using a medium-pressure mercury lamp(U W/cm, $\lambda_{max}=365\;nm)$. In this work, the effects of molecular weights of polyol and diol with low molecular weight into polymer chain on mechanical and dynamic mechanical properties of UV-cured polyurethane acrylates were studied. The structure and properties of the films obtained from the UV photopolymerization of urethane-acrylate prepolymer were investigated by FT-IR spectroscopy, dynamic mechanical measurement, tensile testing, and X-ray diffractometry.