• Communications for Statistical Applications and Methods
• /
• 제9권1호
• /
• pp.87-100
• /
• 2002

Poorly-posed problems in the balanced discriminant analysis was considered. We restrict consideration to the case of observations and the number of variables are the same and small. When these problems exist, we do not use a maximum likelihood estimates(MLE) to estimate covariance matrices. Instead of MLE, an alternative estimate for the covariance matrices are proposed. This alternative method make good use of two regularization parameters, $\lambda$} and $\gamma$. A new test rule for the discriminant function is suggested and examined via limited hut informative simulation study. From the simulation study, it is shown that the suggested test rule gives better test result than other previously suggested method in terms of error rate criterion.

• 한국산림과학회지
• /
• 제80권1호
• /
• pp.9-19
• /
• 1991

본(本) 연구(硏究)에서는 삼림경영계획기간(森林經營計劃期間) 10년 동안의 임분(林分) 축적변화태(蓄積變化態)를 예측(豫測)하기 위하여 Simulation Model이 개발(開發)되고 실제임분(實際林分)에 적용(適用)된다. 강원도(江原道) 인제군(麟蹄郡) 기린면(麒麟面) 진동(鎭東)2리(里) 소재 중부영림서(中部營林署) 인제영리소(麟蹄營理所) 관할 국유림(國有林) 3.844ha를 대상으로 개발(開發)된 Simulation Model에 의해 임분축적변동상황(林分蓄積變動狀況)이 추정(推定)된다. 이때 삼림경영계획자(森林經營計劃者)의 벌채계획(伐採計劃)에 따라 두개의 가정하(假定下)에 Simulation Model 두개가 반복(反復) 실행(實行)된다. 10년간의 생장(生長)과 수확(收穫)을 추정(推定)하기 위해 각 수종별(樹種別) 생장방정식(生長方程式)이 유도(誘導)된다. Simulation Model 1과 2, 두 모델에 의해 예측(豫測)된 10년 동안의 임분축적상황(林分蓄積狀況)을 보면 10-14%의 차이를 보이고 있다. 이것은 Simulation Model 구성(構成)의 중요성(重要性)을 시사(示唆)해 주며 삼림경영계획자(森林經營計劃者)의 의사결정(意思決定)을 통해 잔재(殘在) 임목축적(林木蓄積)에 미치는 영향(影響)이 크다는 것을 보여 준다.

• 한국컴퓨터그래픽스학회논문지
• /
• 제26권5호
• /
• pp.1-13
• /
• 2020

본 논문에서는 흐르는 물에 의해 빙결 시뮬레이션 되어 방향성이 있는 얼음 형태를 안정적으로 모델링 할 수 있는 새로운 방법을 제시한다. 제안하는 얼음 모델링 프레임워크는 빙결 시뮬레이션에서 중요한 얼음의 성장 방향에 점성이 있는 유체의 흐름을 고려한다. 물 시뮬레이션 해법은 암시적 비압축성 유체 시뮬레이션에 새로운 점성 기법을 적용한 방법을 이용하고, 얼음의 방향과 글레이즈(Glaze) 효과는 제안하는 비등방성한 빙결 해법을 이용한다. 물 입자가 얼음 입자로 상태변화하는 조건은 습도와 물의 흐름에 따른 새로운 에너지 함수에 따라 계산된다. 습도는 오브젝트 표면의 가상 수막(Virtual water film)으로 근사되며, 유체의 흐름은 얼음의 성장 방향을 가이드하기 위해 우리의 비등방성한 빙결 해법에 통합된다. 결과적으로 점성이 있는 물의 흐름 방향에 따라 글레이즈와 방향성 있는 빙결 시뮬레이션 결과를 안정적으로 보여준다.

• 한국식품영양과학회지
• /
• 제26권4호
• /
• pp.704-713
• /
• 1997

Simulation의 결과와 PAF식으로 계산된 성장기의 단백질 축적량과의 상대적 차이는 8.8%를 보여 본 연구의 타당성이 확인되었다. 성장률을 나타내는 총 체단백질량에 대한 단백질 축적량은 양질의 단백질을 섭취 할 경우 성장함에 따라 parabolic 형태를 띄나, 사료단백질의 질이 낮아짐에 따라 150일 simulation 기간내에 이러한 형태는 사라졌다. 성장하는 동안 단백질 회전의 두요소가 항상 병행하여 증가하였다 단백질 합성량과 분해량은 서로 병행하여 단백질 질에 따라 차이를 보였는데, 질이 높은 단백질을 섭취하였을 때 단백질 축적의 증가는 높은 단백질 합성량과 분해량에 의해 일어남 을 보여 주었다. 이는 근육단백질의 대사형태(5)를 반영하며, 그리하여 돼지의 총 단백질대사는 근육단백질에 의해 결정됨을 확인해 주었다. 그리고 단백질 질이 낮을 경우 매일의 단백질 대사율과 성장률이 저하할 뿐 아니라 성장지연도 일어남을 보여 주었다 본 연구에서 추정된 필수아미노산의 총 필요량은 28.1g이며, 일반적으로 돼지 사료의 제1 제한 아미노산인 lysine의 필요량은 4.2g이다. 이러한 아미노산 필요량은 NRC에서 권장하는 균형잡힌 아미노산 조성과 유사하다. 이와 같이 본 연구는 동위원소를 사용한 실험을 시행할 필요없이 simulation을 통해 다양한 질의 사료단백질에 따른 단백질 축적량에 대한 자료만으로 성장기의 단백질 회전률과 단백질대사의 동적행위를 나타내어 식이의 단백질 질의 변화에 적응하는 기전을 나타낼 수 있음을 보여주었다.

• Korean Chemical Engineering Research
• /
• 제62권1호
• /
• pp.112-117
• /
• 2024

Numerical simulations on the onset and the growth of viscous fingering during the miscible displacement due to the radial source flow were conducted. With introduction of a new stability criterion, the critical log-viscosity ratio, R_c, was found as a function of the Peclet number, Pe. Similar to the previous linear stability analyses, Pe made the system unstable, i.e., accelerated the onset of instability. For a large Pe system, the present numerical simulation yielded much stable results than the previous theoretical predictions This discrepancy was commonly encountered in the comparison between the theoretical prediction and the experimental finding. Additionally, the difference between the rectilinear system and the present one was also discussed. The present system was found more insensitive to the Peclet number than the rectilinear one.

• Journal of Nutrition and Health
• /
• 제26권9호
• /
• pp.1033-1048
• /
• 1993

This study was designed to describe the effect of the protein quality at different intake level of protein on the protein metabolism in the whole body of growing pigs with a simulation model. Varying to the protein level in feeds, four simulations were conducted. The feed protein level, represented as proportions of digestible protein to the metabolic energy (DP/ME, g/MJ), were 6-8, 11-13, 17-19, and 23-25 DP/ME, respectively. Two protein quality and six weeks of growth time were used at each simulation. The objective function for the simulations was protein deposition in the whole body, which was calculated from the experimental results. The parameters in the simulation were determined by the parameter estimation technique. The results obtained from the simulation were as follows: The protein synthesis and breakdown rates(g/day) in the whole body was increased with the increase of protein quality only at lower or required level of protein intake. They showed a parallel behavior in the course of growth, irrespective of quality and level of feed protein intake. The simulated protein deposition and protein synthesis showed a linear relationship between them at different protein quality and level. The affinity parameter showed a linear relationship between them at different protein quality and level. The affinity parameter showed that arginine, tryptophan and isoleucine were more efficient in the stimulation ofbody protein synthesis. Lysine and phenylalanine+tyrosine were less efficient. The oxidation parameter showed that histidine, pheyalanine+tyrosine were less efficient. The oxidation parameter showed that histidine, phenyalanine+tyrosine, and methionine+cystine were oxidized in larger magnitude than lysine and threonine. The oxidation parameter of most amino acids increased with the increase of protein intake beyond the requirement level, but not any more at highest protein intake level. Finally it was found that the improvement of feed protein quality at the lower or required level of protein intake increase protein deposition through a parallel increase of protein synthesis and breakdown.

• 소성∙가공
• /
• 제14권4호
• /
• pp.384-391
• /
• 2005

The nickel-based alloy Nimonic 80A possesses the excellent strength, and the resistance against corrosion, creep and oxidation at high temperature. Its products are used in aerospace engineering, marine engineering and power generation, etc. Control of forging parameters such as strain, strain rate, temperature and holding time is important because change of the microstructure in hot working affects the mechanical properties. Change of the microstructure evolves by recovery, recrystallization and grain growth phenomena. The dynamic recrystallization evolution has been studied in the temperature range of $950\~1250^{\circ}C$ and strain rate range of $0.05\~5s^{-1}$ using hot compression tests. The metadynamic recrystallization and grain growth evolution has been studied in the temperature range of $950\~1250^{\circ}C$ and strain rate range $0.05,\;5s^{-1}$, holding time range of 5, 10, 100, 600 sec using hot compression tests. Modeling equations are proposed to represent the flow curve, recrystallized grain size, recrystallized fraction and grain growth phenomena by various tests. Parameters in modeling equations are expressed as a function of the Zener-Hollomon parameter. The modeling equation for grain growth is expressed as a function of the initial grain size and holding time. The modeling equations developed were combined with thermo-viscoplastic finite element modeling to predict the microstructure change evolution during hot forging process. The grain size predicted from FE simulation results is compared with results obtained in field product.

• Journal of Welding and Joining
• /
• 제16권5호
• /
• pp.56-66
• /
• 1998

During solidification or welding of alloys, the solute redistribution brings out microsegregation. The microsegregation causes the formation of non-equilibrium second phases, shrinkage and porosity degrading mechanical/chemical properties Therefore, it has been required to predict microsegregation quantitatively. To predict the degree of microsegregation, more exact and appropriate computer simulation technique has been actively used during last two decades. To predict the degree of microsegregation in weld metal, an advanced two dimensional model was suggested. In the new model, both primary and secondary arm regions were defined for the analysis region. The growth in the primary arm regina was assumed to be a planar for effective calculation. Especially, for the growth of a secondary arm, a simple and effective mathematical function was established to show the growing pattern, the solute diffusion in the solid phase was calculated by finite difference method (FDM). The solid-liquid interface movement was considered to be in local equilibrium state. The experiments for welding of 310S stainless steel were carried out in order to examined the reasonability and feasibility of this model. The concentration profiles of the solute predicted by this model were compared with those obtained from experimental works.

• 융합신호처리학회논문지
• /
• 제11권1호
• /
• pp.80-87
• /
• 2010

함수합성법은 주어진 입출력 데이터 쌍으로부터 입출력관계를 충족하는 함수를 예측하는 것으로, 특성을 알 수 없는 시스템을 제어할 때에 필수적이다. 일반적으로 시스템은 비선형인 성질을 갖는 경우가 많고, 함수 합성에 취급하는 변수, 정수, 제약 등으로 조합된 문제가 발생하기가 쉽다. 그 함수를 합성하는 방법 중 한 가지로 유전적 프로그래밍이 제안되고 있다. 이것은 함수를 트리구조로 표시한 함수 트리에 유전적 조작을 적용하여, 입출력 관계를 충족하는 함수 트리를 탐색하는 방법이다. 본 논문에서는 기존의 유전적 프로그래밍에 의한 함수 합성법의 문제점을 지적하고, 새로운 4종류의 개선법을 제안한다. 즉, 함수 트리를 탐색할 때에 함수가 복잡하게 되는 것을 방지하기 위하여 함수 트리의 성장 억제, 조기 수렴을 목표로 하는 국소 탐색법의 채택, 함수 트리 내의 필요 없이 길어지는 요소의 효과적인 삭제, 대상으로 하는 문제의 특성을 이용하는 방법이다. 이러한 개선법을 이용할 경우, 기존의 유전적 프로그래밍에 의한 함수 합성법보다도 짧은 시간에 우수한 구조의 함수 트리가 구해지는 것을 2-spirals 문제에 대하여 컴퓨터 시뮬레이션을 통하여 확인하였다.

• 한국안전학회지
• /
• 제21권4호
• /
• pp.42-48
• /
• 2006

The strategies of maintenance and operation are usually established based on the number of flaws and their size distribution obtained from nondestructive inspection in order to preserve safety of the plant. But non destructive inspection results are different from the physical flaws which really exist in the equipments. In case of a single inspection, it is easy to estimate the number of physical flaws using the POD curve. However, we may be faced with some difficulties in obtaining the number of physical flaws from the periodic in-service non destructive inspection data. In this study a simple method for estimating the number of physical flaws from periodic in-service nondestructive inspection data was proposed. In order to obtain the flaw growth history, the flaw growth was simulated using the Monte Carlo method and the flaw size and the corresponding POD value were obtained for each flaw at each periodic inspection time. The flaw growth rate used in the simulation was statistically calculated from the in-service inspection data. By repeating the simulation numerous flaw growth data could be generated and the effective POD curve was obtained as a function of flaw size. From the effective POD curve the number of physical flaws was obtained. The usefulness and convenience of the proposed method was evaluated from several applications and satisfactory results were obtained.