• Nuclear Engineering and Technology
• /
• 제52권2호
• /
• pp.287-295
• /
• 2020

Group method of data handling (GMDH) is considered one of the earliest deep learning methods. Deep learning gained additional interest in today's applications due to its capability to handle complex and high dimensional problems. In this study, multi-layer GMDH networks are used to perform uncertainty quantification (UQ) and sensitivity analysis (SA) of nuclear reactor simulations. GMDH is utilized as a surrogate/metamodel to replace high fidelity computer models with cheap-to-evaluate surrogate models, which facilitate UQ and SA tasks (e.g. variance decomposition, uncertainty propagation, etc.). GMDH performance is validated through two UQ applications in reactor simulations: (1) low dimensional input space (two-phase flow in a reactor channel), and (2) high dimensional space (8-group homogenized cross-sections). In both applications, GMDH networks show very good performance with small mean absolute and squared errors as well as high accuracy in capturing the target variance. GMDH is utilized afterward to perform UQ tasks such as variance decomposition through Sobol indices, and GMDH-based uncertainty propagation with large number of samples. GMDH performance is also compared to other surrogates including Gaussian processes and polynomial chaos expansions. The comparison shows that GMDH has competitive performance with the other methods for the low dimensional problem, and reliable performance for the high dimensional problem.

• 한국자동차공학회논문집
• /
• 제18권3호
• /
• pp.60-68
• /
• 2010

In order to understand the flow and filtration characteristics in a wall-flow type DPF(Diesel Particulate Filter), 0-D, 1-D, and 3-D simulations are preformed. In this paper, three model are explained and validated with each other. Based on the comparisons with 1-D and 3-D results for the steady state solution, 3-D CFD analysis is preferable to 1-D for the prediction of wall velocity at the inlet and exit plane. Because PM loading process is transient state phenomena, the combination of full 3-D and time dependent simulation is crucial for the configuration of wall channels. New coupling technique, which is the connection between calculated permeability from 0-D lumped parameter model and UDF(User Defined Functions) of main solver, is proposed for the realisti

• International Journal of Aeronautical and Space Sciences
• /
• 제18권1호
• /
• pp.28-37
• /
• 2017

Recent modeling of thermal nonequilibrium processes in simple molecules like hydrogen and nitrogen has indicated that rotational nonequilibrium becomes as important as vibrational nonequilibrium at high temperatures. In the present work, in order to analyze rovibrational nonequilibrium, the rotational mode is separated from the translational-rotational mode that is usually considered as an equilibrium mode in two- and multi-temperature models. Then, the translational, rotational, and electron-electronic-vibrational modes are considered separately in describing the thermochemical nonequilibrium of nitrogen behind a strong normal shock wave. The energy transfer for each energy mode is described by recently evaluated relaxation time parameters including the rotational-to-vibrational energy transfer. One-dimensional post-normal shock flow equations are constructed with these thermochemical models, and post-normal shock flow calculations are performed for the conditions of existing shock-tube experiments. In comparisons with the experimental measurements, it is shown that the present thermochemical model is able to describe the rotational and electron-electronic-vibrational relaxation processes of nitrogen behind a strong shock wave.

• 한국연소학회지
• /
• 제15권3호
• /
• pp.67-73
• /
• 2010

The temperature distribution of diesel particulate filter with five partitioned electric heaters is numerically analyzed to investigate the condition of regenerating ceramic filter. The commercial code STAR-$CCM+^{(R)}$ is utilized to simulate multi-dimensional steady hot air flow in DPF. In order to verify the computational results, thermocouples are used to measure the temperature distribution in DPF. Computational results agree well with experimental ones. The results show that the maximum temperature in DPF is lowered as the mass flow rate of exhaust gas increases, which means that the more power in heater will be necessary as the engine speed increases. Compared with heater placed at center, heater at circumference has the higher maximum temperature in DPF. The maldistribution of flow field in front of heater has the main influence on the temperature distribution in DPF.

• 대한기계학회논문집B
• /
• 제28권1호
• /
• pp.72-80
• /
• 2004

Steady flow of an ER (electro-rheological) fluid in a two-dimensional electrode channel is studied by using FEM. Hydrodynamic interactions between the particles and the fluid are calculated by solving the Navier-Stokes equation combined with the equation of motion for each particle, where the multi-body electrostatic interaction is described by using point-dipole model. Motion of the particles in the ER fluid is elucidated in conjunction with the mechanisms of the flow resistance and the increase of viscosity. The ER effects have been studied by varying the Mason number and volume fraction of particles. These parameters have an influence on the formation of the chains resulting in the changes of the fluid velocity and the effective viscosity of ER fluids.

• 대한기계학회:학술대회논문집
• /
• 대한기계학회 2003년도 춘계학술대회
• /
• pp.1821-1826
• /
• 2003

In this study, we discuss CIP method, which can treat compressible/incompressible problem and multi-phase problem. We can apply this method to the general equations by using CCUP. In this paper, non-staggered grid arrangement and predictor-corrector method are used to enhance later development and the solution accuracy and convergence performance. To validate the numerical algorithm proposed in this paper, the two-dimensional unsteady flow in the driven cavity is simulated. The numerical results of this subject using the present algorithm are compared with other numerical results. As a result, it is prived that the present scheme gives stable and improved solutions, and the results even coarse grid are in good agreement with other result using a fine grid arrangement.

• 대한조선학회논문집
• /
• 제47권1호
• /
• pp.1-10
• /
• 2010

In the present paper, a numerical method based on the constraint interpolation profile (CIP) method is applied for simulating two-dimensional violent sloshing problems. The free surface boundary value problem is considered as a multiphase problem which includes water and air. A stationary Cartesian grid system is adopted, and an interface capturing method is used to trace the shape of free surface profile. The CIP combined unified procedure (CCUP) scheme is applied for flow solver, and the tangent of hyperbola for interface capturing (THINC) scheme is used for interface capturing. Numerical simulations have been carried out for partially-filled 2D tanks under forced sway and roll motions at various filling depths and frequencies. The computational results are compared with experiments and/or the other numerical results to validate the present numerical method.

• 대한설비공학회:학술대회논문집
• /
• 대한설비공학회 2008년도 동계학술발표대회 논문집
• /
• pp.475-480
• /
• 2008

A pre/post processor program based GUI(Graphic User Interface) by using the MFC and OpenGL library in the Windows OS have been developed for the analysis of the passenger flow. Using this program, users are able to generate and modify the meshes of multi-storied subway station, set all the parameters for the solver, and obtain the results of the simulation such as transient passenger motions and passenger streak lines in 3-dimensional graphic view.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2004년도 추계학술대회논문집
• /
• pp.170-177
• /
• 2004

A single degree of freedom system and transmissibility are key concepts in many problems of vibration isolation. In order to apply this approach, however, several assumptions must be satisfied, which are often not realistic. In this paper, an approach using vibration power flow is introduced to deal with vibration isolations in a more practical way. Procedures of this approach and some results of research are presented. Difficulties in this method are also discussed.

• 에너지공학
• /
• 제24권3호
• /
• pp.122-127
• /
• 2015

본 연구에서는 클린룸에서의 3차원 유동해석을 수행하였다. 로봇실험 장치에서 로봇의 움직임에 의해 형성되는 유동장에 대한 수치적인 분석을 본 연구에서는 다루고 있으며, 클린룸에서 로봇의 움직임에 대한 영향을 분진이 쌓이는 영역을 찾기 위해서 진행된다. 다관절 로봇의 움직임에 의해 생성되는 바닥 표면의 분진은 유동장 분석에 의해 예측될 수 있으며, 해석결과는 커다란 스월유동이 움직이는 로봇 주위에서 생성된다고 분석된다. 따라서, 가장 바람직한 유동 조건은 공기의 입구와 출구부의 위치를 다시 정하여 유동속도를 변화시키면서 검토해야 한다.