• Proceedings of the Computational Structural Engineering Institute Conference
• /
• 2011.04a
• /
• pp.752-755
• /
• 2011

이종재료의 열전달문제 수치해석시 추가적으로 만족시켜야 하는 계면경계조건들의 존재와 계면경계로 인한 불연속면의 처리는 근사함수의 구성 뿐만 아니라 수치기법의 개발 자체를 어렵게 만든다. 본 논문에서는 계면경계의 불연속성을 모델링하는 특수한 함수를 포함하고 계면경계조건을 항상 만족시킬 수 있는 근사함수를 구성하고, 계면경계문제의 강형식을 직접 이산화하며 고속으로 해를 계산할 수 있는 이동최소제곱 차분법을 제시한다. 계면경계조건이 매입된 이동최소제곱 차분법으로 이종재료의 열전달문제를 해석한 결과, 높은 정확성과 효율성을 갖는 것을 확인할 수 있었다.

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.13 no.2
• /
• pp.100-108
• /
• 2001

Along the shoreline a special treatment is required to simulate movement of periodic waves such as tsunami and tide because of continuous movement of shoreline as waves rise and recede. A moving boundary treatment is first proposed to track the movement of shoreline in this study. The treatment is then employed to obtain a maximum inundation area to be used for mitigation of coastal flooding. The obtained maximum inundation zone for a specific location is compared to that of available observed data. A reasonable agreement is observed.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2011.05a
• /
• pp.65-65
• /
• 2011

하천의 하구지역은 유역의 보전, 개발 논의가 매우 다양하게 전개되고 있는 지역으로 개발계획들로 인한 환경변화로 유출 특성 및 유사이동 변화에 따른 정밀한 조사연구를 통한 대처방안을 수립이 필요한 지역이다. 본 논문은 2차원 수치해석이 가능하고 복잡한 해안구조를 포함한 하구지역의 흐름해석이 가능한 MIKE21 FM 프로그램을 통하여 홍수기시의 조위조건변화에 따른 수치해석을 수행하여 자연하구역인 한강하구지역을 중심으로 수위 및 유사이동의 변화 특성을 분석하였다. 첨두유량 발생 시점과 하류경계조건의 변화에 따른 하구지역의 수위변화를 살펴보면, 대상 지역의 하천구간에 해당하는 전류수위표 지점에서 기수역 수로 지점에 비하여 변동폭이 가장 크게 나타났으며, 하류의 외해와 인접할수록 변동폭이 줄어드는 것으로 나타났다. 또한, 대상지역의 상류구간은 하천유량의 영향이 크기 때문에 상류구간일수록 창조시간에 비하여 낙조시간이 길게 분포되는 것으로 나타난다. 유사이동의 특성은 하천의 유속이 크게 감소하는 부분에 퇴적이 발생되며 강화북수로의 단면이 급격하게 변화하는 석모도와 교동도사이의 수로, 염하수로의 입구부분에서는 퇴적층이 발달하는 것으로 나타났다. 따라서, 한강하구와 같이 하구둑이 설치되지 않은 하천의 하구지역에 대해서는 상류와 하류 경계조건의 특성에 따라서 복잡한 수리학적 특성이 발생하기 때문에 개발이나 구조물 축조에 앞서 더욱 많은 현장조사와 특성분석을 통한 보다 정확한 경계조건의 구축이 필요할 것으로 사료된다.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.15 no.1
• /
• pp.61-68
• /
• 1991

The system such as railway bridge can be modelled as the restrained beam with intermediate supports. This kind of structures are subject to the moving load, which has a great effect on dynamic stresses and can cause sever motions, especially at high velocities. Therefore, to analyze the dynamic characteristics of the system due to the moving load is very important. In this paper, the governing equation of motion of a restrained beam subjected to the moving load is derived by using the Hamilton's principle. The orthogonal polynomial functions, which are trial functions and satisfying the geometric and dynamic boundary conditions, are obtained through simple procedure. The dynamic response of the system subjected to the moving loads is obtained by using the Galerkin's method and the numerical time integration technique. The numerical tests for various constraint, velocity and boundary conditions were preformed. Furthermore, the effects of mass of the moving load are studied in detail.

• Journal of the Korean Society for Railway
• /
• v.19 no.1
• /
• pp.54-66
• /
• 2016

In this study, the absorbing boundary condition was adopted to prevent elastic wave propagation due to abrupt load condition changes in moving force analysis. It was proven to be effective in reducing inappropriate noise components. Furthermore, to verify whether fixed point excitation in a laboratory test can simulate train movement effectively, the results of dynamic analysis for fixed point excitation and moving force conditions were analyzed. The dynamic stability of an asphalt track structure under a KTX train condition was also studied. Additionally, the dynamic safety of asphalt tracks was verified by comparing the results of a moving force analysis under a KTX train load and a quasi-static analysis under the design standard Korean train load.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.32 no.4
• /
• pp.215-222
• /
• 2019

The heat conduction problem with discontinuous material coefficients generally consists of the conservative equation, boundary condition, and interface condition, which should be additionally satisfied in the solution procedure. This feature often makes the development of new numerical schemes difficult as it induces a layered singularity in the solution fields; thus, a special approximation is required to capture the singular behavior. In addition to the approximation, the construction of a total system of equations is challenging. In this study, a wedge function is devised for enriching the approximation, and the interface condition itself is embedded in the moving least squares(MLS) derivative approximation to consistently satisfy the interface condition. The heat conduction problem is then discretized in a strong form using the developed derivative approximation, which is named as the interface immersed MLS difference method. This method is able to efficiently provide a numerical solution for such interface problems avoiding both numerical quadrature as well as extra difference equations related to the interface condition enforcement. Numerical experiments proved that the developed numerical method was highly accurate and computationally efficient at solving the heat conduction problem with interfacial jump as well as the problem with a geometrically induced interfacial singularity.

• Journal of the KSME
• /
• v.24 no.4
• /
• pp.275-283
• /
• 1984

유한요소기법은 원만한 온도변화를 갖는 이동경계문제에 대해 적용할 수 있음을 설명하였다. 상변화가 일어나는 요소에서 어느 한 시간 단계에 변화할 수 있는 온도의 범위는 상변화 온도 구간 보다 적어야 원만한 수치해를 얻을 수 있음을 소개하였다. 잠열의 효과를 엔탈피의 온도에 대한 변화율로 처리하여 열용량 행렬을 계산할 수 있음을 설명하였다. 급격한 변화를 갖는 온 도분포를 수치적으로 근사해를 구할 때 나타나는 파동형상은 연속적인 변수를 부분적으로 근사 해를 가정 하는데서 포함되는 오차에 기인한다. 따라서 이러한 급격한 변화의 변수를 취급하는 기법이 요구된다. 상변화를 수반하는 이동경계문제는 물의 결빙금속의 열처리, 용접, 고체연료의 연소 등에서 많이 접하게 되나 전문적인 프로그램이 없으므로(저자가 조사한 바에 의하면) 온 도분포를 요구하는 문제에서는 이의 선결이 요망되고 있다. 특히 구조해석문제에서 시간에 따라 온도조건이 변화는 경우가 많아 온도분포에 대해 따로 계산하고 그 후 구조해석을 하는 방법이 있으나, 이는 상변화와 같은 물성의 큰 변화가 없는 경우에만 가능하다. 따라서 상변화가 있는 문제에서 열탄성(thermo-elasticity)이나 열탄소성(thermoelasto-plasticity)해석을 하기 위해서는 이동경계와 구조해석을 동시에 하여야 한다.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.26 no.1
• /
• pp.39-48
• /
• 2013

This paper develops a 2-D moving least squares(MLS) difference method for Stefan problem by extending the 1-D version of the conventional method. Unlike to 1-D interfacial modeling, the complex topology change in 2-D domain due to arbitrarily moving boundary is successfully modelled. The MLS derivative approximation that drives the kinetics of moving boundary is derived while the strong merit of MLS Difference Method that utilizes only nodal computation is effectively conserved. The governing equations are differentiated by an implicit scheme for achieving numerical stability and the moving boundary is updated by an explicit scheme for maximizing numerical efficiency. Numerical experiments prove that the MLS Difference Method shows very good accuracy and efficiency in solving complex 2-D Stefan problems.

• Journal of Korea Water Resources Association
• /
• v.36 no.6
• /
• pp.937-947
• /
• 2003

A shoreline, which has no the water depth, moves continuously as waves rise up and recede. Therefore, a special boundary treatment is required to track properly the movements of the shoreline in numerical modeling of the behavior of tsunamis or tides near a coastal zone. In this study, convective terms in nonlinear shallow-water equations are discretized explicitly by using a second-order upwind scheme to describe a moving shoreline more accurately. An oscillatory flow motion in a circular paraboloidal basin has been employed to validate the performance of the developed numerical model. Computed results of instantaneous free surface displacements are compared with those of analytical solutions and existing numerical solutions. The run-up heights in the vicinity of a circular island have also been calculated and obtained numerical results have been shown against available laboratory measurements. A good agreement has been observed.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.9 no.6
• /
• pp.813-823
• /
• 1998

This paper analyzed the circularly polarized conformal microstrip patch antennal using the unsplit anisotropic perfectly matched layer(UAPML) method. Also, this paper are treated effectively the edge and corner parts on the 3 dimensional UAPML. Especially, to analyze microstrip patch antennas with the coaxial feeder line, it was applied to mixed the UAPML with Mur's first order absorbing boundary condition. Therefore this paper suggest the new the method to mix the UAPML with Mur's first order absorbing boundary condition. The results show the time responses of electromagnetics $E_z$ and $H\chi'$, input impedances of coaxial cable and radiation patterns of strip parchs on the single and the array patchs with central frequencies 1.575 GHz, 1.778 GHz and 4.8 GHz in L-band and C-band for mobile communication. The results of this paper shows that its results was compared the Mur's first order abc and mixed the second order dispersion boundary condition(SDBC) with the Mur's first order absorbing boundary condition. In accordance with, the validity of the method is confirmed.