• Proceedings of the Korea Water Resources Association Conference
• /
• 2023.05a
• /
• pp.294-294
• /
• 2023

보나 댐에 설치된 배수관이나 지하에 매설된 상하수도관과 같이 물을 운반하기 위한 관로 뿐만 아니라 유류를 운송하는 관, LPG와 같은 기체를 운송하는 관처럼 유체를 운송할 때 다양한 관로를 사용한다. 그 중 사용범위나 제작에 대한 용이성 등에 의해서 원형관이 주로 사용된다. 따라서 위와 같은 원형관 내의 관수로 흐름분포에 대한 연구는 아주 중요하며 필요하다. 원형관 내 흐름분포는 관에 연결된 펌프 혹은 수조나 저수지의 수위에 의한 관내 압력에 의해 지배되어지며, 관 내부 표면의 거칠기나 관의 꺾인 정도 등 다양한 요인에 영향을 받는다. 본 연구에서는 일반적인 실험결과를 도출하기 위하여 직선의 원형관을 대상으로 실험과 수치모의를 동시에 수행한다. 실험은 높이 3m, 길이 4.5m, 폭 1.5m 수조의 바닥에서부터 0.45m 위에 위치한 1.8m 길이의 아크릴 재질의 원형관이 설치된 수조에서 진행되었으며, 수치모의 또한 동일한 규모에서 수행되었다. 수조의 수위를 변화시켜 여러가지 레이놀즈 수에 따른 관 내 흐름의 변화에 대하여 분석하는 것이 목적이며, 실험결과와 수치모의 계산결과 간의 비교검증을 통해 분석한다. 이 연구에서는 난류의 영향을 RANS(Reynolds-averaged Navier-Stokes)와 LES(large eddy simulation)을 혼합한 형태인 Hybrid RANS/LES 모형 중 하나인 DES(detacged-eddy simulation)기법을 이용하여 해석함으로써, 빠른 유속으로 인한 관 내의 강한 와(eddy)에 대한 효과를 예측하는데 한계가 있는 RANS 모형과 벽면근처에서의 흐름 해석을 위해 굉장히 높은 격자해상도가 요구되어 계산적 비용 측면에서 효율이 떨어지는 LES 모형의 한계를 극복하고자 한다. DES 모형은 벽에서 떨어진 와에 대하여 LES로 직접해석하고, 벽 근처에서의 흐름에 대해선 난류모형을 통해 모델링함으로써 벽 근처 계산격자와 계산량을 줄이면서 LES와 비슷한 결과를 얻을 수 있다. 실험결과와 수치모의 계산결과 사이의 비교검증을 통하여 관 내 흐름에 대한 수치모의의 적용성을 평가하고 실험에서 측정하기 어려운 난류강도와 압력변동의 상세한 특성을 수치모의를 통해 분석함으로서 관 내 흐름특성에 대하여 자세히 분석하였다.

• Journal of the Society of Naval Architects of Korea
• /
• v.29 no.1
• /
• pp.43-54
• /
• 1992

A method of evaluating the velocity components induced by a Havelock source is presented in this paper. The mathematical manipulation of x, y and z-derivative of the Green function of the Havelock source by the use of contour integration on the complex plane has resulted in the expressions that can be readily incorporated with computer program. The accuracy and efficiency that can be secured by the use of the present mathematical expressions have been convincingly found to be highly satisfactory.

• 한국방재학회:학술대회논문집
• /
• 2010.02a
• /
• pp.52.2-52.2
• /
• 2010

본 연구에서는 자유수면의 계산을 위해 동수압을 해석하는 수치모형을 제시하였다. 동수압과 자유수면을 고려하기 위해, 비점성 Navier-Stokes 방정식을 3단계로 나누어 해석하였다. 제1단계에서는 동수압과 자유수면을 전번 시간단계에서 계산된 값으로 대입하여 차분하였으며 이 차분식은 NGMRES(Newton-Generalized Minimal Residual) 방법을 이용해 음해적으로 해석되었다. 이때 계산된 유속장은 연속방정식을 고려하지 않았으므로 각 계산격자에서 질량보존법칙을 만족하지 않을 수도 있다. 제2단계에서 유속과 동수압 보정항으로 이루어진 식을 연속방정식에 대입하여 얻어지는 타원형 방정식인 동수압-포와송 방정식을 해석하므로 여기서 얻어지는 유속은 질량보존법칙을 만족하게 된다. 마지막 3 단계에서는 자유수면과 최종 유속을 계산하였다. 새로 개발된 수치모형을 검증하기위해 정사각형 탱크에서 수면의 자유 진동 문제에 적용한 결과 수치해는 해석해와 잘 일치하였다.

• Journal of the Society of Naval Architects of Korea
• /
• v.38 no.3
• /
• pp.1-13
• /
• 2001

RANS solver has been developed to solve the flows past a ship with a drift angle. The solver employs a finite volume method for the spatial discretization and Euler implicit method for the time integration. Turbulent flows are simulated by Spalart-Allmaras one-equation model. Developed solver is applied to analyze the hydrodynamic forces and flows of two tankers with a same forebody but different afterbodies. The computed flows and hydrodynamic forces are compared with the measured flows and captive model test data. The computed results show good agreements with experimental data and show clearly the effects of stern hull form on the hydrodynamic forces and the flows.

• Journal of Korea Water Resources Association
• /
• v.44 no.5
• /
• pp.363-376
• /
• 2011

1-D hydrodynamic numerical models have been most widely used in the field of flood analysis. The model's input data are upstream/downstream boundaries, roughness coefficients, cross-sections, and so on, and computational distance step and time step are the most important factors in order to guarantee the computational accuracy, stability, and efficiency. In this study, a theoretical explanation is presented for the basis of the previous empirical selection criteria of cross-section's location; also, the estimation technique of computationally variable distance step is proposed to reflect the properties of flow at every computational time step. Combining this technique with 1-D unsteady numerical model, it was applied to two events of Teton dam failure flood and the Han River flood. The numerical experimental results demonstrate that the accuracy and stability is increased when used more interpolated cross-sections and show that the proposed technique of computationally variable distance step has the same order of accuracy with smaller numbers of cross-section than previous empirical selection criteria. The practical use of this technique will be possible to analyze the river floods with high efficiency as well as accuracy and stability.

• Proceedings of the Korean Society of Coastal and Ocean Engineers Conference
• /
• 1998.09a
• /
• pp.106-116
• /
• 1998

Hasselmann(1963b)은 Neumann 스펙트럼에 대한 비선형에너지 전달함수의 수치계산을 처음으로 행하였다. 그 후 Sell and Hassetmann(1972)은 동일한 계산법을 JONSWAP 스펙트럼에 적용하여 비선형에너지 전달에 의하여 스펙트럼의 첨두가 저주파 영역으로 천이하는 현상을 발견하였다. 그러나 이 계산법은 방대한 시간을 필요로 하며 또한 수치적인 불안정성 때문에 계산 정도가 떨어지는 단점을 갖고 있다. (중략)

• Journal of the Society of Naval Architects of Korea
• /
• v.35 no.1
• /
• pp.1-14
• /
• 1998

A two-dimensional numerical model for tidal currents based on the depth averaged equation is developed. The mode1 employs a rectangular grid system for its simplicity in the application of complicate coastal shore lines. To raise computing efficiency, implicit approximate factorization scheme is implemented in solving governing equations. An upwind-differencing is used to discretize convective terms, which provides a numerical dissipation automatically and suppresses any oscillations caused by nonlinear instabilities. Some numerical tests are made against the analytic solutions of a linearized shallow water equation to validate the developed numerical scheme, and comparisons of the model prediction with the analytic solution are satisfactory. As a real application, the tidal currents are computed on the Inchon area where the tidal currents are important for the design of new canal which is under construction.

• Korean Journal of Optics and Photonics
• /
• v.20 no.6
• /
• pp.375-380
• /
• 2009

The scattering of a normally incident monochromatic light by optically active concentric double cylinders is studied by using a numerical method. A numerical code is developed on the basis of the analytical solutions, and the angular distributions of the intensity and the polarization of the scattered light are computed for some specific cases. The numerical code can be used to non-invasively determine the size, structure, and composition of a double cylinder with high accuracy by measuring the angular distribution of the scattered light by an experiment.

• Journal of Wetlands Research
• /
• v.14 no.4
• /
• pp.439-455
• /
• 2012

Coupled wave induced current modeling system(WIC) was developed from combining with the nearshore spectral wave model, SWAN, the wave induced force model, WIF, and the flow model, EFDC. The reasonable results were obtained from WIC modeling system. The ratio of the wave height calculated with respect to refraction and diffraction effects over submerged spherical shoal was occurred approximately 1~5 % errors compared to Goda(2000)'s result. The radiation stress suggested by Longuet-Higgins and Stewart(1960), the stresses due to rollers in breaking waves proposed by Dally and Osiecki(1994), and Kim(2004)'s new spreading approach instead of the previous lateral mixing approach were added to calculate wave induced force. The results of the WIC modeling system show good agreement with Nishimura et al.(1985)'s laboratory measurements and better than Kim(2004)'s 2 dimensional depth averaged numerical computations for a plane beach with detached breakwater. The present flow field computed agrees reasonably well with the measured flow field. The relative merit of WIF model in WIC modeling system is unconditional stable for time increment.

• Explosives and Blasting
• /
• v.36 no.3
• /
• pp.10-18
• /
• 2018

The dynamic behaviour of the rock mass under the dynamic load is different from the static application of the maximum load of the same size. An experimental approach to investigating rock behavior under dynamic loads is more difficult than that under static conditions in control of dynamic loads, measurement and analysis of the results. Numerical methods are less constrained by performing the experiments numerically, rather than experimental ones, so they can be very powerful analytical tool at the design stage. However, even if the algorithms of the analysis method are appropriate, careful analysis is required because the calculation results may vary largely depending on input data and boundary conditions. In this paper, when investigating the behavior of rock structures under dynamic load numerically, the effects of boundary conditions, dynamic load and calculation time step, and dynamic load characteristics on the calculation results were reviewed to provide guidance on setting up boundary conditions and calculation time step related to dynamic analysis.