• Nuclear Engineering and Technology
• /
• 제49권2호
• /
• pp.434-441
• /
• 2017

Containment venting is one of several essential measures to protect the integrity of the final barrier of a nuclear reactor during severe accidents, by which the uncontrollable release of fission products can be avoided. The authors seek to develop an optimization approach to venting operations, from a simulation-based perspective, using an integrated severe accident code, THALES2/KICHE. The effectiveness of the containment-venting strategies needs to be verified via numerical simulations based on various settings of the venting conditions. The number of iterations, however, needs to be controlled to avoid cumbersome computational burden of integrated codes. Bayesian optimization is an efficient global optimization approach. By using a Gaussian process regression, a surrogate model of the "black-box" code is constructed. It can be updated simultaneously whenever new simulation results are acquired. With predictions via the surrogate model, upcoming locations of the most probable optimum can be revealed. The sampling procedure is adaptive. Compared with the case of pure random searches, the number of code queries is largely reduced for the optimum finding. One typical severe accident scenario of a boiling water reactor is chosen as an example. The research demonstrates the applicability of the Bayesian optimization approach to the design and establishment of containment-venting strategies during severe accidents.

• Structural Engineering and Mechanics
• /
• 제69권4호
• /
• pp.361-369
• /
• 2019

The determination of Paris' law parameters based on crack growth experiments is an important procedure of fatigue life assessment. However, it is a challenging task because it involves various sources of uncertainty. This paper proposes a novel probabilistic method, termed the S-N Paris law (SNPL) method, to quantify the uncertainties underlying the Paris' law parameters, by finding the best estimates of their statistical parameters from the S-N curve data using a Bayesian approach. Through a series of steps, the SNPL method determines the statistical parameters (e.g., mean and standard deviation) of the Paris' law parameters that will maximize the likelihood of observing the given S-N data. Because the SNPL method is based on a Bayesian approach, the prior statistical parameters can be updated when additional S-N test data are available. Thus, information on the Paris' law parameters can be obtained with greater reliability. The proposed method is tested by applying it to S-N curves of 40H steel and 20G steel, and the corresponding analysis results are in good agreement with the experimental observations.

• Animal Bioscience
• /
• 제34권7호
• /
• pp.1100-1104
• /
• 2021

Objective: The objective of present study was to estimate heritability of non-return rate (NRR) and success of first insemination (SFI) by using the Bayesian approach with Gibbs sampling. Methods: Heifer Traits were denoted as NRR-h and SFI-h, and cow traits as NRR-c and SFI-c. The variance covariance components were estimated using threshold model under Bayesian procedures THRGIBBS1F90. Results: The SFI was more relevant to evaluating success of insemination because a high percentage of animals that demonstrated no return did not successfully conceive in NRR. Estimated heritability of NRR and SFI in heifers were 0.032 and 0.039 and the corresponding estimates for cows were 0.020 and 0.027. The model showed low values of Geweke (p-value ranging between 0.012 and 0.018) and a low Monte Carlo chain error, indicating that the amount of a posteriori for the heritability estimate was valid for binary traits. Genetic correlation between the same traits among heifers and cows by using the two-trait threshold model were low, 0.485 and 0.591 for NRR and SFI, respectively. High genetic correlations were observed between NRR-h and SFI-h (0.922) and between NRR-c and SFI-c (0.954). Conclusion: SFI showed slightly higher heritability than NRR but the two traits are genetically correlated. Based on this result, both two could be used for early indicator for evaluate the capacity of cows to conceive.

• 한국수자원학회논문집
• /
• 제47권9호
• /
• pp.753-766
• /
• 2014

본 논문에서는 기후변화로 인한 한반도 주요 권역에서의 미래 평균해수면 상승을 장기 조위자료를 사용하여 통계적으로 추정하는 연구를 수행하였다. 먼저 5개 조위 관측소로부터 얻어진 장기 조위자료에 대한 비모수적 경향성 검정인 Mann-Kendall 검정을 통해 관측된 자료의 경향성을 검정하였으며, 이를 보다 정량적으로 분석하기 위하여 Bayesian 변동점 분석 기법을 적용하였다. 특히 이 연구에서는 4개의 미래 평균해수면 상승 시나리오와 5개 관측소의 지역별 평균해수면 상승 자료를 결합시키기 위하여 변동점 분석결과를 활용하였다. 제안된 절차는 미래 평균해수면 상승 시나리오의 시작년도를 결정함에 있어 18.6년의 주기를 사용하지 않고 변동점 분석결과를 사용함으로써, 지역적 특성을 효과적으로 반영할 수 있도록 개선되었다. 변동점 분석결과를 사용하여 한반도의 권역별 미래 해수면상승을 분석한 결과, 제주 권역(제주 조위관측소)이 가장 뚜렷한 해수면 상승을 나타냈다. 서해안 권역(보령 조위관측소)과 남해안 권역(부산 조위관측소)에서는 두 번째로 높은 해수면 상승의 증가가 추정되었으며, 마지막으로 남해안 권역(여수 조위관측소)와 동해안 권역(속초 조위관측소)에서 가장 낮은 해수면 상승의 증가가 추정되었다.

• 콘크리트학회논문집
• /
• 제22권4호
• /
• pp.499-510
• /
• 2010

이 연구의 목적은 Bayesian 통계법을 통하여 얻어진 만족도 곡선을 활용하여 체계적으로 콘크리트 재료성능을 평가하고 배합설계를 하는 것이다. 단일변수 만족도 곡선은 콘크리트 성능기준을 만족할 확률을 콘크리트 재료변수 함수로서 나타낸다. 여러 개의 만족도 곡선을 결합해 하나의 만족도 곡선으로 나타내기 위하여 Importance Factor와 Goodness value라는 신규개념을 도입하여 서로 다른 재료변수들이 콘크리트성능에 미치는 영향을 정량화하고 서로 다른 재료변수들을 공통된 하나의 변수로 통합하는 것을 가능하도록 하였다. 또한 PBMD 과정에 의한 설계예제를 제시함으로써 목표지향적 콘크리트배합설계의 한 방법을 제시하고 그 유효성에 대해 증명하였다. 마지막으로, 실제 구조물에 대한 적용 가능성을 확인하기 위해 PBMD과정에 의한 콘크리트의 기대성능 결과값과 ACI 기준에 의한 결과값을 비교하였다.

• Communications for Statistical Applications and Methods
• /
• 제3권1호
• /
• pp.155-168
• /
• 1996

In this paper we consider the multiprocess dynamic normal model with parameters having a time dependent non-linear structure. We develop and study the recursive estimation procedure for the proposed model with normality assumption. It turns out thst the proposed model has nice properties such as insensitivity to outliers and quick reaction to abrupt changes of pattern.

• Journal of the Korean Statistical Society
• /
• 제27권3호
• /
• pp.331-348
• /
• 1998

In this paper, we have proposed some robust Bayes estimators using ML-II priors as well as certain empirical Bayes estimators in estimating the finite population mean in the presence of auxiliary information. These estimators are compared with the classical ratio estimator and a subjective Bayes estimator utilizing the auxiliary information in terms of "posterior robustness" and "procedure robustness" Also, we have addressed the issue of choice of sampling design from a robust Bayesian viewpoint.

• 한국안전학회지
• /
• 제18권4호
• /
• pp.164-168
• /
• 2003

Bayes theorem, suggested by the British Mathematician Bayes (18th century), enables the prior estimate of the probability of an event under the condition given by a specific This theorem has been frequently used to revise the failure probability of a component or system. 2-Stage Bayesian procedure was firstly published by Shultis et al. (1981) and Kaplan (1983), and was further developed based on the studies of Hora & Iman (1990) Papazpgolou et al., Porn(1993). For a small observed failure number (below 12), the estimated reliability of a system or component is not reliable. In the case in which the reliability data of the corresponding system or component can be found in a generic reliability reference book, however, a reliable estimation of the failure probability can be realized by using Bayes theorem, which jointly makes use of the observed data (specific data) and the data found in reference book (generic data).

• 한국수산과학회지
• /
• 제41권2호
• /
• pp.107-112
• /
• 2008

To evaluate uncertainty and risk in biological reference points, we applied a bootstrapping method and a Bayesian procedure to estimate the related confidence intervals. Here we provide an example of the maximum sustainable yield (MSY) of turban shell, Batillus cornutus, estimated by the Schaefer and Fox models. Fitting the time series of catch and effort from 1968 to 2006 showed that the Fox model performs better than the Schaefer model. The estimated MSY and its bootstrap percentile confidence interval (CI) at ${\alpha}=0.05$ were 1,680 (1,420-1,950) tons for the Fox model and 2,170 (1,860-2,500) tons for the Schaefer model. The CIs estimated by the Bayesian approach gave similar ranges: 1,710 (1,450-2,000) tons for the Fox model and 2,230 (1,760-2,930) tons for the Schaefer model. Because uncertainty in effort and catch data is believed to be greater for earlier years, we evaluated the influence of sequentially excluding old data points by varying the first year of the time series from 1968 to 1992 to run 'backward' bootstrap resampling. The results showed that the means and upper 2.5% confidence limit (CL) of MSY varied greatly depending on the first year chosen whereas the lower 2.5% CL was robust against the arbitrary selection of data, especially for the Schaefer model. We demonstrated that the bootstrap and Bayesian approach could be useful in precautionary fisheries management, and we advise that the lower 2.5% CL derived by the Fox model is robust and a better biological reference point for the turban shells of Jeju Island.

• 대한산업공학회지
• /
• 제16권2호
• /
• pp.135-147
• /
• 1990

A nuclear power plant can be viewed as a large complex man-machine system where high system reliability is obtained by ensuring that sub-systems are designed to operate at a very high level of performance. The chance of severe accident involving at least partial core-melt is very low but once it happens the consequence is very catastrophic. The prediction of risk in low probability, high-risk incidents must be examined in the contest of general engineering knowledge and operational experience. Engineering knowledge forms part of the prior information that must be quantified and then updated by statistical evidence gathered from operational experience. Recently, Bayesian procedures have been used to estimate rate of accident and to predict future risks. The Bayesian procedure has advantages in that it efficiently incorporates experts opinions and, if properly applied, it adaptively updates the model parameters such as the rate or probability of accidents. But at the same time it has the disadvantages of computational complexity. The predictive distribution for the time to next incident can not always be expected to end up with a nice closed form even with conjugate priors. Thus we often encounter a numerical integration problem with high dimensions to obtain a predictive distribution, which is practically unsolvable for a model that involves many parameters. In order to circumvent this difficulty, we propose a method of approximation that essentially breaks down a problem involving many integrations into several repetitive steps so that each step involves only a small number of integrations.