• 대한화학회지
• /
• 제37권5호
• /
• pp.527-536
• /
• 1993

Chitin을 강알칼리로 탈아세틸화시켜 합성한 chitosan을 증류수에 팽윤시킨 다음 글리세린을 가하여 교반하였다. 이 고분자 용액에 약물인 silver sulfadiazine을 가하여 경피흡수용 고분자 matrix을 제조하였다. 이렇게 제조된 고분자 matrix로부터 약물의 방출거동과 고분자 matrix 변수와의 상관관계 등을 조사함으로써 지속적이고 조절된 경피흡수제형으로서의 사용 가능성과 특성을 조사하였다. 고분자 matrix 내의 약물의 함유량과 matrix의 두께가 증가할수록 약물의 방출시간은 더 지연되었다. 그러나 글리세린의 함유량이 증가함에 따라 약물의 방출시간은 오히려 감소하였다. 약물의 함유량, 글리세린의 함유량 및 matrix의 두께가 증가할수록 겉보기 방출속도상수(K)값도 증가하였다.이상과 같이 chitosan은 의약의 방출조절형제제로서 가능성을 나타냈으며, 약물로 사용된 silver sulfadiazine의 방출거동은 Higuchi model에 따른 확산으로 생각되었다.

• 대한기계학회논문집A
• /
• 제21권2호
• /
• pp.317-326
• /
• 1997

An investigation was performed to study the matrix cracking and delamination in laminated composite plates due to transverse impact. A model was developed for predicting the initiation of the matrix cracking and the shape and size of impact-induced delamination in laminated composite plates resulting from the ballistic impact. The model consists of a stress analysis and a failure analysis. A transient finite element analysis which was based on the higher-order shear deformation theory was adopted for calculating the stresses inside the laminated composite plates during impact. A failure analysis was used to predict the initial intraply matrix cracking and the shape and size of the interface delamination in the laminates. As a results, a shear matrix cracking which was governed by the transverse interlaminar shear stress occured at the middle layer near the midplane of laminates and a bending matrix cracking which was governed by the transverse inplane stress occured at the bottom layer near the surface of laminates. In a thick laminates, a shear matrix cracking generated first at the middle layer of laminates, but in a thin laminates, a bending matrix cracking generated first at the bottom layer of laminates.

• 대한교통학회지
• /
• 제22권6호
• /
• pp.57-66
• /
• 2004

관측교통량으로부터 기종점 OD 행렬을 추정할 때 적정한 관측구간을 선정한 후 선정된 구간의 교통량을 이용하여 기종점 OD행렬을 추정하는 것이 기종점 ODGODFUF의 추정력을 향상시키고 경제적인 효율성(관측비용 감소)을 제고하기 위한 방법이라고 판단하였다. 본 연구에서는 관측구간을 선정하기 위해 정수계획법을 이용하여 관측구간을 선정하는 모형식을 제시하였으며 이 기법을 이용하여 교통량 관측구간을 선정한 후, 성정된 관측구간의 교통량으로부터 기종점 OD행렬을 추정하였다. 또한 적정한 관측구간으로 선정된 교통량으로부터 추정된 기종점 OD행렬의 추정력을 비교분석하기 위하여 임의적으로 관측구간을 선정하여 기종점 OD행렬을 추정한 기존의 연구결과와 비교하였다. 비교분석한 결과, 적정한 관측구간을 선정하여 추정한 구간교통량 및 추정 기종점 OD행렬의 추정력이 임의적으로 관측구간의 교통량을 이용한 결과의 오차보다 더 개선되는 것으로 분석되었다. 따라서 관측교통량으로 기종점 OD행렬을 추정하기 위해서는 적정한 관측구간을 선정하여 선정된 구간의 교통량으로부터 기종점 OD행렬을 추정하는 것이 오차를 줄일수 있으며 추정력을 향상시킬수 있는 방법이라고 하겠다.

• Communications for Statistical Applications and Methods
• /
• 제18권1호
• /
• pp.79-87
• /
• 2011

공간적, 시간적으로 퍼져나가는 전염성이 강한 질병인 수두자료를 이용하여 공간 시계열 자료를 분석하는데 있어 일반적으로 알려진 ARIMA 모형에 적합하여 분석을 행하면 공간적인 정보를 반영하지 못하기 때문에 기존에 시간만을 고려한 시계열 분석방법에 공간통계의 공간적 정보를 반영한 공간시계열 모형을 고려한다. 공간시계열 모형에서 공간의 위치 및 영향은 시계열 모형에 공간적 정보로써 가중치행렬을 더 함으로써 처리 가능해진다. 가중치행렬은 지리적으로 인접한 지역일수록 공간의존도가 높다는 것을 반영한 것이며 공간시계열 모형의 연구에서 가중치행렬은 인접한 지역들은 동일한 영향을 줄 것이라 가정하였다. 따라서 본 논문에서는 공간시계열 모형인 STARMA 모형과 STBL 모형에 대한 식별방법, 통계적 추론 및 예측력 비교에 대해 연구하였고 특히, 모수추정의 알고리즘 비교와 공간시계열 모형의 예측력 비교를 통해 Kalman-Filter 방법의 우수성을 보이고자 한다.

• 대한수학회보
• /
• 제56권1호
• /
• pp.187-200
• /
• 2019

Truncated Hankel operators are compressions of classical Hankel operators to model spaces. In this paper we describe matrix representations of truncated Hankel operators on finite-dimensional model spaces. We then show that the obtained descriptions hold also for some infinite-dimensional cases.

• 한국복합재료학회:학술대회논문집
• /
• 한국복합재료학회 2002년도 춘계학술발표대회 논문집
• /
• pp.207-210
• /
• 2002

Metal matrix composites(MMCs) are increasingly attractive for high technology components such as aerospace applications and transportations due to their high strength, stiffness, and toughness. Many processes for fabricating MMCs have been developed, and relatively simple Foil-Fiber-Foil method is usually employed in solid state consolidation processes. During the consolidation processes at high temperature, densification occurs by the inelastic flow of the matrix materials, and the process is coupled with the conditions of pressure, temperature and volume fraction of fiber and matrix materials. This is particularly important in titanium matrix composites, and thus a generic model based on micro-mechanical approaches enabling the evolution of density over time to be predicted has been developed. The mode developed is then implemented into FEM so that practical process simulation has been carried out. Further the experimental investigation of the consolidation behavior of SiC/Ti-6Al-4V composites using vacuum hot pressing has been performed, and the results obtained are compared with the model predictions.

• 한국복합재료학회:학술대회논문집
• /
• 한국복합재료학회 2002년도 춘계학술발표대회 논문집
• /
• pp.203-206
• /
• 2002

The present study proposes a theoretical model for predicting the matrix crack density growth of each layer in composite laminates subjected to thermo-mechanical loads. Each layer with matrix cracks is treated as an equivalent continuum of degraded elastic stiffnesses which are functions of the matrix crack density in each slyer. The energy release rate as a function of the degraded elastic stiffnesses is then calculated for each layer as functions of thermo-mechanical loads externally applied to the laminate. The matrix crack densities of each layer in general laminates are predicted as functions of the thermo-mechanical loads applied to a number of laminates. Comparisons of the present study with experimental data in the open literatures are also provided.

• Journal of Electrical Engineering and Technology
• /
• 제8권5호
• /
• pp.1138-1145
• /
• 2013

A novel predictive current control strategy for a sparse matrix converter is presented. The sparse matrix converter is functionally-equivalent to the direct matrix converter but has a reduced number of switches. The predictive current control uses a model of the system to predict the future value of the load current and generates the reference voltage vector that minimizes a given cost function so that space vector modulation is achieved. The results show that the proposed controller for sparse matrix converters controls the load current very effectively and performs very well through simulation and experimental results.

• 말소리와 음성과학
• /
• 제1권1호
• /
• pp.69-73
• /
• 2009

This paper proposes an efficient global covariance-based principal component analysis (GCPCA) for speaker identification. Principal component analysis (PCA) is a feature extraction method which reduces the dimension of the feature vectors and the correlation among the feature vectors by projecting the original feature space into a small subspace through a transformation. However, it requires a larger amount of training data when performing PCA to find the eigenvalue and eigenvector matrix using the full covariance matrix by each speaker. The proposed method first calculates the global covariance matrix using training data of all speakers. It then finds the eigenvalue matrix and the corresponding eigenvector matrix from the global covariance matrix. Compared to conventional PCA and Gaussian mixture model (GMM) methods, the proposed method shows better performance while requiring less storage space and complexity in speaker identification.

• Communications for Statistical Applications and Methods
• /
• 제25권1호
• /
• pp.61-70
• /
• 2018

Modeling of the random effects covariance matrix in generalized linear mixed models (GLMMs) is an issue in analysis of longitudinal categorical data because the covariance matrix can be high-dimensional and its estimate must satisfy positive-definiteness. To satisfy these constraints, we consider the autoregressive and moving average Cholesky decomposition (ARMACD) to model the covariance matrix. The ARMACD creates a more flexible decomposition of the covariance matrix that provides generalized autoregressive parameters, generalized moving average parameters, and innovation variances. In this paper, we analyze longitudinal count data with overdispersion using GLMMs. We propose negative binomial loglinear mixed models to analyze longitudinal count data and we also present modeling of the random effects covariance matrix using the ARMACD. Epilepsy data are analyzed using our proposed model.