• Computational Structural Engineering
• /
• v.8 no.4
• /
• pp.65-74
• /
• 1995

Direct method developed for the inelastic analysis of planar frames subjected to monotonic loads is extended to cyclic loads. Two frame elements for Direct Method(inelastic truss and inelastic beam) are developed. The accuracy and reliability of the preposed method is verified by comparing the analysis results of example with step-by-step analysis. Direct Method is superior to Step-by-step analysis in view of reliability of solution and analysis cost.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2004.10a
• /
• pp.347-350
• /
• 2004

In this paper, the acceleration is studied for the rigid-plastic FEM of metal forming simulation. In the FEM, the direct iteration and Newton-Raphson iteration are applied to obtain the initial solution and accurate solution respectively. In general, the acceleration scheme for the direct iteration is not used. In this paper, an Aitken accelerator is applied to the direct iteration. In the modified Newton-Raphson iteration, the step length or the deceleration coefficient is used for the fast and robust convergence. The step length can be determined by using the accelerator. The numerical experiments have been performed for the comparisons. The faster convergence is obtained with the acceleration in the direct and Newton-Raphson iterations.

• Proceedings of the ESK Conference
• /
• 1996.04a
• /
• pp.307-312
• /
• 1996

최대허용하중을 결정에는 여러가지 방법이 있지만, 그 중 특히 심리육체적 방법은 빈도수가 적고 반복적인 들기 작업을 평가하기 위해서 널리 사용되어 왔다. 하지만 이 방법은 작업방법의 조합이 많아질수록 최대허용하중을 결정하는데 걸리는 시간은 커지게 된다. 이를 단축시키 위한 한가지 대안인 직접추정법은 일종의 비율 척도 (ratio scale)를 사용하여 자극과 주관적 평가를 연결시키는 방법으로서 넓은 범위의 자극에 대한 인간의 반응을 자극에 짧게 노출되더라도 정확 하게 예측할 수 있는 정확한 수단인 것으로 보여진 바 있다. 이에 본 연구에서는 네 가지의 들기 빈도 (분당 1, 2, 4, 6 회)와 두 가지의 수직 이동거리 (Floor to Knuckle, Knuclke to Shoulder)로 구성되는 8가지의 대칭적인 (symmetrical)들기 작업에 대하여 심리육체적 방법과 기준작업을 달리한 직접추정법을 사용하여 최대허용하중을 결정하고 그 값들을 비교, 분석하였으며 이를 근거로 직접 추정법의 기준변화에 따른 추정능력 정도에 대하여 검토하였다.

• Journal of the Korean Geotechnical Society
• /
• v.20 no.2
• /
• pp.15-26
• /
• 2004

Based on an estimating method for post-cyclic strength and stiffness with cyclic triaxial tests proposed by one of the authors, cyclic Direct Simple Shear (DSS) tests were carried out to confirm whether the method can be adapted to DSS test on fine-grained soils: silty clay, plastic silt, and non-plastic silt. Results from cyclic and post-cyclic DSS tests were interpreted by a modified method as adopted for cyclic and post-cyclic triaxial tests. In particular, influence of plasticity index for fine-grained soils and initial static shear stress (ISSS) was emphasised. Findings obtained from the present study are: (i) liquefaction strength ratio of fine-grained soils decreases with decreasing plasticity index and increasing ISSS; (ii) plasticity index and ISSS did not markedly influence relation between equivalent cyclic stiffness and shear strain relations; (iii) the higher the plasticity index of fine-grained soils is, the less the strength ratio decreases with increment of a normalcies excess pore water pressure (NEPWP); (iv) stiffness ratio of plastic silt has large activity decrease rapidly with increasing excess pore water pressure; and (v) post-cyclic strength and stiffness results from DSS tests agree well with those predicted by the method modified from a procedure used for triaxial test results.

• Geophysics and Geophysical Exploration
• /
• v.10 no.2
• /
• pp.147-153
• /
• 2007

The conjugate gradient (CG) method is one of the most efficient algorithms for solving a linear system of equations. In addition to being used as a linear equation solver, it can be applied to a least-squares problem. When the CG method is applied to large-scale three-dimensional inversion of magnetotelluric data, two approaches have been pursued; one is the linear CG inversion in which each step of the Gauss-Newton iteration is incompletely solved using a truncated CG technique, and the other is referred to as the nonlinear CG inversion in which CG is directly applied to the minimization of objective functional for a nonlinear inverse problem. In each procedure we only need to compute the effect of the sensitivity matrix or its transpose multiplying an arbitrary vector, significantly reducing the computational requirements needed to do large-scale inversion.

• Journal of the Earthquake Engineering Society of Korea
• /
• v.9 no.5 s.45
• /
• pp.75-86
• /
• 2005

In performance-based design methods, it is clear that the evaluation of the nonlinear response is required. Analysis methods available to the design engineer today are nonlinear time history analyses, or monotonic static nonlinear analyses, or equivalent static analyses with simulated inelastic influences. The nonlinear time analysis is the most accurate method in computing the nonlinear response of structures, but it is time-consuming and necessitate more efforts. Some codes proposed the capacity spectrum method based on the nonlinear static analysis to determine earthquake-induced demand. The nonlinear direct spectrum method is proposed and studied to evaluate nonlinear response of structures, without iterative computations, given by the structural linear vibration period and yield strength from pushover analysis. The purpose of this paper is to compare the accuracy and the reliability of approximate nonlinear methods with respect to shear buildings and various earthquakes. The conclusions of this study are summarized as follows: 1) Linear capacity spectrum method may fail to find a convergent answer or make a divergence. Even if a convergent answer is found, it has a large error in some cases and the error varies greatly depending on earthquakes. 2) Although nonlinear capacity spectrum method need much less calculation than capacity spectrum method and find an answer in any case, it may be difficult to obtain an accurate answer and generally large error occurs. 3) The nonlinear direct spectrum method is thought to have good applicability because it produce relatively correct answer than other methods directly from pushover curves and nonlinear response spectrums without additional and iterative calculations.

• Journal of the Earthquake Engineering Society of Korea
• /
• v.8 no.1
• /
• pp.17-27
• /
• 2004

A new earthquake design method performing iterative calculations using secant stiffness was developed. The proposed design method has the advantages of convenience and stability in numerical analysis because it uses elastic analysis. At the same time, the proposed design method can accurately estimate the strength and ductility demands on the members because it performs the analysis on the inelastic behavior of structure using iterative calculation. In the present study, the procedure of the proposed design method was established, and a computer program incorporating the proposed method was developed. Design examples using the proposed method were presented, and its advantages were presented by the comparisons with existing design methods using elastic or inelastic analysis. The proposed design method, as an integrated method of analysis and design, can address the earthquake design strategy devised by the engineer. such as ductility limit on each member, the design concept of strong column - weak beam, and etc. In addition, through iterative calculations on the structure preliminarily designed only with member sizing, the strength and ductility demands of each member can be directly calculated so as to satisfy the given design strategy. As the result. economical and safe design can be achieved.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.12 no.4
• /
• pp.642-651
• /
• 1988

The present paper develops finite element procedures to calculate displacements, strains and stresses in initially stressed elastic solids subjected to static or time-dependent loading conditions. As a point of departure, we employ Hamilton's principle to obtain nonlinear equations of motion characterizing the displacement in a solid. The equations of motion reduce to linear equations of motion if incremental stresses are assumed to be infinitesimal. In the case of linear problem, finite element solutions are obtained by Newmark's direct integration method and by modal analysis. An analytic solution is referred to compare with the linear finite element solution. In the case of nonlinear problem, finite element solutions are obtained by Newton-Raphson iteration method and compared with the linear solution. Finally, the effect of the order of Gauss-Legendre numerical integration on the nonlinear finite element solution, has been investigated.

• Journal of Advanced Navigation Technology
• /
• v.13 no.1
• /
• pp.62-67
• /
• 2009

The demand for high performance computer grows to solve large linear systems of equations in such engineering fields - circuit simulation for VLSI design, image processing, structural engineering, aerodynamics, etc. Many various parallel processing systems have been proposed and manufactured to satisfy the demand. The properties of linear system determine what algorithm is proper to solve the problem. Direct methods or iterative methods can be used for solving the problem. In this paper, an intelligent parallel iterative method for solving linear systems of equations with large sparse matrices is proposed and its efficiency is proved through simulation.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 1995.04a
• /
• pp.591-595
• /
• 1995

진동구조물의 해석 또는 설계변경 등을 위해서는 구조물의 강체특성(질량, 질량중심위치, 관성모멘트, 관성적, 주축방향)을 정확히 추정할 필요가 있다. 본 연구에서는 Direct System Identification Method를 애용하여 진동 시험으로부터 얻어진 잡음이 혼입된 데이터로 부터 특성행렬(질량행렬, 감쇄행렬, 강성행렬의총칭)을 직접 규명하는 반복계산에 의한 노이즈에 강인한 방법의 개발과 가상중심점을 이용하지 않고 실험으로 부터 얻어지는 병진성분 데이터를 직접 이용하여 강체특성을 추정하는 방법을 개발한다. 마운트 지지된 평판모델에 대한 실험적 응용과 시뮬레이션에 의한 soild 모델 응용으로 본 연구의방법의 타당성을 검증한다.