• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.20 no.4
• /
• pp.493-499
• /
• 2007

The Taylor expansion expresses a differentiable function and its coefficients provide good approximations for the given function and its derivatives. In this study, m-th order Taylor Polynomial is constructed and the coefficients are computed by the Moving Least Squares method. The coefficients are applied to the governing partial differential equation for solid problems including crack problems. The discrete system of difference equations are set up based on the concept of point collocation. The developed method effectively overcomes the shortcomings of the finite difference method which is dependent of the grid structure and has no approximation function, and the Galerkin-based meshfree method which involves time-consuming integration of weak form and differentiation of the shape function and cumbersome treatment of essential boundary.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.22 no.6
• /
• pp.735-740
• /
• 2012

In this paper, we introduce design methodologies of polynomial radial basis function neural network classifier with the aid of Principal Component Analysis(PCA) and Linear Discriminant Analysis(LDA). By minimizing the information loss of given data, Feature data is obtained through preprocessing of PCA and LDA and then this data is used as input data of RBFNNs. The hidden layer of RBFNNs is built up by Fuzzy C-Mean(FCM) clustering algorithm instead of receptive fields and linear polynomial function is used as connection weights between hidden and output layer. In order to design optimized classifier, the structural and parametric values such as the number of eigenvectors of PCA and LDA, and fuzzification coefficient of FCM algorithm are optimized by Artificial Bee Colony(ABC) optimization algorithm. The proposed classifier is applied to some machine learning datasets and its result is compared with some other classifiers.

• Review of KIISC
• /
• v.2 no.4
• /
• pp.37-47
• /
• 1992

이 논문에서 우리는 여러 가지 분할 항등식을 유도했고 제한된 분할에 관한 새로운 항등식을 증명하고, 분할의 기본적인 이론과 분할함수(Partition Number Function)가 다항식 함수가 아니라는 것을 보이며, n의 분할의 수 p(n)에 대한 ㅎ계(Lower Bound)를 얻기 위해 Stirling의 n !에 대한 근사값을 소개한다. 그리고Hardy-Ramanujan 공식, Euler 항등식과 p(n)의 순환식을 유도하며, 그리고 d$_{m}$ (n)이 n을 m개의 부분으로 분할하는 분할의 수를 나타낼 때 우리는 d$_{m}$ (n) 에 관한 일반적인 공식을 p(n)과 함께 행렬식의 형태로 표현한다.

• Proceedings of the KIEE Conference
• /
• 2005.07d
• /
• pp.2793-2795
• /
• 2005

고장검출식별시스템의 성능을 좌우하는 가장 중요한 요소는 잔차발생알고리즘이다. 잔차발생기구는 측정된 입출력변수 정보에 근거하여 설계되며 고장의 검출 및 식별이 가능하도록 감결합(decoupling) 특성을 가져야 한다. 본 논문에서는 전달함수 접근 방법 중 최근 제안된 최소차 다항식 기저(minimal polynomial basis)를 이용한 잔차발생기구 설계기법을 소개하고, 이를 VTOL의 센서 및 구동장치고장 검출에 적용하여 그 성능을 검토하였다.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.23 no.6
• /
• pp.533-538
• /
• 2013

The damage caused by Recent frequently occurring locality torrential rains is increasing rapidly. In case of densely populated metropolitan area, casualties and property damage is a serious due to landslides and debris flows and floods. Therefore, the importance of predictions about the torrential is increasing. Precipitation characteristic of the bad weather in Korea is divided into typhoons and torrential rains. This seems to vary depending on the duration and area. Rainfall is difficult to predict because regional precipitation is large volatility and nonlinear. In this paper, Very short-term precipitation forecasting pattern model is implemented using KLAPS data used by Korea Meteorological Administration. we designed very short term precipitation forecasting pattern model using GA-based RBFNNs. the structural and parametric values such as the number of Inputs, polynomial type,number of fcm cluster, and fuzzification coefficient are optimized by GA optimization algorithm.

• Journal of the Korean Data and Information Science Society
• /
• v.20 no.6
• /
• pp.1015-1027
• /
• 2009

Many problems of confounding and identifiability for polynomial models in an experimental design can be solved using methods of algebraic geometry. The theory of $Gr\ddot{o}bner$ basis is used to characterize the design. In addition, a fractional factorial design can be uniquely represented by a polynomial indicator function. $Gr\ddot{o}bner$ bases and indicator functions are powerful computational tools to deal with ideals of fractions based on each different theoretical aspects. The problem posed here is to give how to move from one representation to the other. For a given fractional factorial design, the indicator function can be computed from the generating equations in the $Gr\ddot{o}bner$ basis. The theory is tested using some fractional factorial designs aided by a modern computational algebra package CoCoA.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.31 no.3A
• /
• pp.321-327
• /
• 2006

In this paper, when a H-polarized plane wave is incident on the grating consisting of uniform resistive strips, electromagnetic scattering is analyzed using the moment of methods (MoM). The current density of each resistive strip on a grounded dielectric plane is fixed by zero at both edges. To satisfy the condition at both ends of each resistive strip, the induced surface current density is expanded in a series of cosine and sine functions. The scattered electromagnetic fields are expanded in a series of floquet mode functions. The boundary conditions are applied to obtain the unknown current coefficients. According to the variation of the involving parameters such as strip width and spacing and angle of the incident field, numerical simulations are performed by applying the Fourier-Galerkin moment method. The numerical results of the normalized reflected power for resistive strips case for zero and several resistivities are obtained.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.14 no.3
• /
• pp.41-48
• /
• 2004

Finite field multiplication and division are important arithmetic operation in error-correcting codes and cryptosystems. The elements of the finite field GF($2^m$) are represented by bases with a primitive polynomial of degree m over GF(2). We can be easily realized for multiplication or computing multiplicative inverse in GF($2^m$) based on a normal basis representation. The number of product terms of logic function determines a complexity of the Messay-Omura multiplier. A normal basis exists for every finite field. It is not easy to find the optimal normal element for a given primitive polynomial. In this paper, the generating method of normal basis is investigated. The normal bases whose product terms are less than other bases for multiplication in GF($2^m$) are found. For each primitive polynomial, a list of normal elements and number of product terms are presented.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2005.11a
• /
• pp.479-482
• /
• 2005

본 논문에서는 퍼지 뉴럴네트워크의 새로운 구조인 Fuzzy Set-based Polynomial Neural Networks(FSPNN)을 소개한다. 제안된 모델은 일반적인 최적화 방법과 정보 입자를 이용하여 네트워크를 설계한다. 최종 구조는 Fuzzy Set-based Polynomial Neuron(FSPN)을 기반으로 설계한 FPNN과 동일하다. 첫째로 FSPNS의 종합적인 설계방법(유전자 알고리즘을 이용한 최적 구조 탐색)에 대해 소개한다. FSPNN에 관계되는 입력변수의 개수, 후반부 다항식의 차수, 멤버쉽 함수의 수 그리고 입력변수 개수에 따른 입력변수를 유전자 알고리즘을 통하여 동조한다. 두 번째로, 입력 변수의 개별적인 퍼지 규칙 형성과 퍼지 공간 분할 및 삼각형 멤버쉽 함수의 초기 정점을 HCM 클러스터링을 통한 Information Granules로 정의한다. 또한 데이터 입자의 중심을 이용하여 후반부의 구조를 결정한다. 이 네트워크의 성능은 기존에 퍼지 또는 뉴로퍼지 모델링에서 실험된 모델링 표준치를 이용하여 평가한다.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.13 no.9
• /
• pp.937-946
• /
• 2002

In this paper, we present a stable solution of the transient electromagnetic scattering from the conducting objects. This method does not utilize the conventional marching-on in time (MOT) solution. Instead we solve the time domain integral equation by expressing the transient behavior of the induced current in terms of weighted Laguerre polynomials. By using this basis functions for the temporal variation, the time derivative in the integral equation can be handled analytically. Since these temporal basis functions converge to zero as time progresses, the transient response of the induced current does not have a late time oscillation. To show the validity of the proposed method, we solve a time domain electric feld integral equation and compare the results of MOT, Mie solution, and the inverse discrete Fourier transform (IDFT) of the solution obtained in the frequency domain.