• 한국전산구조공학회논문집
• /
• 제18권2호
• /
• pp.103-111
• /
• 2005

본 논문에서는 수치적분 정도를 향상시킬 수 있는 새로운 무요소 기법을 제안한파 저자들에 의해 페트로프-갤러킨 자연요소법(PG-NEM)이라 명명된 이 새로운 기법은 보로노이 다이어그램과 델라우니 삼각화에 기반을 두고 있으며, 이는 BG-NEM이라 불리는 기존의 자연요소법과 개념적으로 동일하다. 하지만, 동일한 시험 형상함수와 시도 형상함수를 선택하는 BG-NEM과는 달리, PG-NEM에서는 지지영역이 적분을 위한 배경격자에 정확하게 일치하도록 시험 형상함수를 독립적으로 선택하는 페트로프-갤러킨 개념에 기반을 두고 있다. 따라서, 제안된 기법은 BG-NEM과 비교하여 수치적분 정도를 현저히 향상시킬 것으로 기대된다.

• Earthquakes and Structures
• /
• 제14권1호
• /
• pp.45-53
• /
• 2018

A family of the structure-dependent methods seems very promising for time integration since it can simultaneously have desired numerical properties, such as unconditional stability, second-order accuracy, explicit formulation and numerical dissipation. However, an unusual overshoot, which is essentially different from that found by Goudreau and Taylor in the transient response, has been experienced in the steady-state response of a high frequency mode. The root cause of this unusual overshoot is analytically explored and then a remedy is successfully developed to eliminate it. As a result, an improved formulation of this family method can be achieved.

• 한국전산구조공학회:학술대회논문집
• /
• 한국전산구조공학회 1993년도 봄 학술발표회논문집
• /
• pp.37-44
• /
• 1993

There are a lot of linear dynamic analysis methods for complex structures. Each method has advantages and shortcomings. Method of dynamic analysis for complex structure is selected considering characteristics of dynamic loading, computer facility available number of degree of freedem and accuracy of results. It is a main point of view to get economical results rather then accurate ones for analysis of general complex structures, Mode superposition method and direct integration method are generally used. However, the characteristics of load is not considered in mode superpositon method, the personal computer cannot be used in direct integration methods. To over-come these shortcomings, the component mode method incorporating Ritz algorithm updated is proposed to solve economically dynamic behavior of the structures. The purpose of study is a formulation of algorithm, and computer programing suitable for dynamic analysis of the complex structure in personal computer environment.

• 대한전기학회논문지:시스템및제어부문D
• /
• 제51권7호
• /
• pp.286-293
• /
• 2002

This paper presents a new method for finding the Block Pulse series coefficients and deriving the Block Pulse integration operational matrices which are necessary for the control fields using the Block Pulse functions. In order to apply the Block Pulse function technique to the problems of continuous-time dynamic systems more efficiently, it is necessary to find the more exact value of the Block Pulse series coefficients and drives the related integration operational matrices by using the Lagrange second order interpolation polynomial.

• 한국건설관리학회논문집
• /
• 제18권3호
• /
• pp.42-51
• /
• 2017

건설 프로젝트의 생산과정에서 성과를 지속적으로 측정하고 평가하는 것은 프로젝트관리의 성공적인 수행을 위한 핵심요소이다. 성과는 일반적으로 목적물에 할당 된 공정과 원가 관련 정보를 정량적으로 평가함으로써 측정되며, 객관적인 성과측정을 위해 공정과 원가가 통합적으로 계획 관리되어야 한다. 본 연구는 공정과 원가의 통합 모델을 통해 성과 측정 방법론을 제안하고 진도관리의 기반을 마련하는데 목적이 있다. 이를 위해 공정계획과 원가계획의 표준화된 업무 프로세스를 개발하였고, EVM과 기존의 통합 모델을 분석하여, 성과측정 유닛을 활용한 WBS와 CBS의 통합 방법론을 제시하였다. 플랜트 프로젝트의 시공단계를 중심으로 공통 성과측정의 대상은 공종으로 선정하여 성과측정 유닛을 규정하고 공정과 비용정보를 연계시켜 성과비율을 산정하였다. 또한 리소스의 적용시 원가요소 중 소모성자원의 비용을 배제하여 진도측정의 객관성을 향상하였다. 본 결과는 실제 플랜트 사례 프로젝트에 적용하여 실증되었으며, 개발된 방법론은 향후 프로젝트의 성과를 보다 객관적인 기준으로 평가할 수 있어 보다 효과적인 진도관리 방법이 될 것으로 판단된다.

• 대한기계학회:학술대회논문집
• /
• 대한기계학회 2001년도 춘계학술대회논문집B
• /
• pp.48-52
• /
• 2001

In this paper presents an accurate AFM used that is free from the Z-directional distortion of a servo actuator is described. Two mathematical correction methods by the in-situ self-calibrationare employed in this AFM. One is the method by the integration, and the other is the method by inverse function of the calibration curve. The in situ self-calibration method by the integration, the derivative of the calibration curve function of the PZT actuator is calculated from the profile measurement data sets which are obtained by repeating measurements after a small Z-directional shift. Input displacement at each sampling point is approximately estimated first by using a straight calibration line. The derivative is integrated with reference to the approximate input to obtain the approximate calibration curve. Then the approximation of the input value of each sampling point is improved using the obtained calibration curve. Next the integral of the derivative is improved using the newly estimated input values. As a result of repeating these improving process, the calibration curve converges to the correct one, and the distortion of the AFM image can be corrected. In the in situ self-calibration through evaluating the inverse function of the calibration curve, the profile measurement data sets were used during the data processing technique. Principles and experimental results of the two methods are presented.

• 한국지진공학회:학술대회논문집
• /
• 한국지진공학회 2006년도 학술발표회 논문집
• /
• pp.113-124
• /
• 2006

Despite the site classification method was improved in the previous study, the response spectrum would be required to be modified by adjusting the integration interval to calculate the site coefficients because the response spectra did not match well the average spectral accelerations obtained by site response analyses in the range of long periods. In this paper, new response spectra for each site categories were determined by adjusting the integration interval of long-period site coefficient $F_{v}$ from $0.4{\sim}2.0$ to $0.4{\sim}1.5$ second. It matched well the average spectral accelerations and new response spectrum, and it was also improved compared to the current site classification system.

• 한국지진공학회논문집
• /
• 제7권5호
• /
• pp.11-17
• /
• 2003

일반적으로 토목구조물에서 계측된 가속도 신호의 대부분은 적분이 매우 힘들다. 그 이유는 토목구조에서 계측된 가속도 신호는 일반적으로 비정상신호이며 또한 비가우시안 노이즈와 저주파 노이즈를 포함하고 있어, 저주파 성분이 증폭되는 적분과정에서 수치적 불안정성이 발생할 수 있기 때문이다. 따라서, 본 논문에서는 비정상 신호처리에 탁월한 웨이블릿 변환의 개념을 비가우시안 노이즈와 저주파 노이즈에 대해 확장한 수정된 웨이블릿 변환을 이용한 가속도 기록의 이중 적분방법을 제시하였다. 또한, 예제해석을 통해 제시된 방법이 비정상 신호의 노이즈 및 비가우시안 노이즈와 저주파 노이즈를 제거에 우수한 성능을 보이고 있음을 보였다.

• Structural Engineering and Mechanics
• /
• 제2권4호
• /
• pp.395-402
• /
• 1994

An improved finite element-transfer matrix method is applied to the transient analysis of plates with large displacement under various excitations. In the present method, the transfer of state vectors from left to right in a combined finite element-transfer matrix method is changed into the transfer of generally incremental stiffness equations of every section from left to right. Furthermore, in this method, the propagation of round-off errors occurring in recursive multiplications of transfer and point matrices is avoided. The Newmark-${\beta}$ method is employed for time integration and the modified Newton-Raphson method for equilibrium iteration in each time step. An ITNONDL-W program based on this method using the IBM-PC/AT microcomputer is developed. Finally numerical examples are presented to demonstrate the accuracy as well as the potential of the proposed method for dynamic large deflection analysis of plates with random boundaries under various excitations.

• 한국언어정보학회:학술대회논문집
• /
• 한국언어정보학회 2007년도 정기학술대회
• /
• pp.481-488
• /
• 2007

In this paper, we propose a method to classify movie review documents into positive or negative opinions. There are several approaches to classify documents. The previous studies, however, used only a single classifier for the classification task. We describe a multiple classifier for the review document classification task. The method consists of three classifiers based on SVMs, ME and score calculation. We apply two voting methods and SVMs to the integration process of single classifiers. The integrated methods improved the accuracy as compared with the three single classifiers. The experimental results show the effectiveness of our method.