• 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.

• 한국측량학회지
• /
• 제40권4호
• /
• pp.325-332
• /
• 2022

우주 측지 기술 사이의 상대적인 위치 관계를 설명하는 벡터를 결정하기 위해서는 VLBI IVP (Very Long Baseline Interferometry Invariant Point)의 위치를 정밀하게 계산하여야 한다. 이를 위해 일반적으로 VLBI 안테나에 반사 타겟을 부착한 후 필라들로부터 경사 거리, 수평각, 수직각을 관측한다. 그 다음 단계에서는 관측값과 미지수를 연결하는 수학 모델을 이용하여 조정 계산을 수행하게 된다. 따라서 계산된 미지수는 관측값의 정밀도에 영향을 받게 된다. 이때 특히 문제가 되는 것은 반사 타켓이 일반적인 측량 정밀도를 확보하기 어려운 곳에 위치하고 있다는 점이다. 즉, 반사 타겟의 방향을 조정하여 측량 기기에 정확하게 맞출 수 없다는 것이다. 따라서 이러한 부분은 관측 오차에 또 다른 형태로 나타날 것이며 조정 계산 시 오차 모델링에 오류를 발생시킬 수도 있다. 본 연구에서는 조정 계산 후 계산된 잔차의 특성에 대한 분석을 수행하였다. 먼저 관측 타입별 통계 분석을 통해 정규성을 검정하였으며 분산에 차이가 있는 지에 대한 검정도 실시하였다. 관측 타입별로 등분산 검정을 한 경우 분산이 서로 다른 것으로 나타났다. 각 필라에 대해 관측 타입별 등분산 검정을 했을 때 경사 거리와 수평 및 수직각 사이에는 분산에 차이가 있는 것으로 나타났다. 따라서 결합 측량으로부터 최적의 결과를 얻기 위해서는 관측 오차에 대해 보다 세분화된 모델링이 필요한 것으로 나타났다.

• 한국측량학회지
• /
• 제19권3호
• /
• pp.237-244
• /
• 2001

본 연구에서는 50년된 흙댐을 대상으로 정적 GPS관측방법을 이용하여 변형 모니터링을 실시하였다. 실험대상은 모르타르로 응급 복구된 지역으로, 기준망 관측은 2차에 걸쳐 기준점들간의 기선벡터를 관측하였고, 모니터링 점들에 대해서는 4회 걸쳐 관측되었다. 측정된 관측값의 분석은 Data snooping법에 의해 과대오차를 제거하고, 우연오차를 최소제곱법으로 조정하여 좌표를 산출하였고, 분산-공분산행렬로 변위를 계산하였다. 또한 조정된 좌표는 95% 신뢰타원으로 정밀도를 표시하였으며, 계산된 변위의 크기와 방향은 그림으로 나타내었다.

• Journal of Positioning, Navigation, and Timing
• /
• 제9권2호
• /
• pp.59-63
• /
• 2020

The antenna arrays are known to be effective for GPS anti-jamming and the performance can be improved further if a dual-polarized antenna array is used. However, when the Minimum Variance Distortionless Response (MVDR) beamformer is used as a signal processing algorithm for the dual-polarized antenna array, the anti-jamming performance can degrade in the presence of errors in the steering vector that is a key factor of the MVDR beamformer. Therefore, in this paper, the effect of the steering vector error on the anti-jamming performance of the dual-polarized antenna array is analyzed by simulations and the result is compared to that of the single-polarized antenna array.

• Journal of the Korean Statistical Society
• /
• 제28권2호
• /
• pp.225-234
• /
• 1999

We consider the following semi-parametric non-linear mixed effect regression model : y\ulcorner=f($\chi$\ulcorner;$\beta$)+$\sigma$$\mu$($\chi$\ulcorner)+$\sigma$$\varepsilon$\ulcorner,i=1,…,n,y*=f($\chi$;$\beta$)+$\sigma$$\mu$($\chi$) where y'=(y\ulcorner,…,y\ulcorner) is a vector of n observations, y* is an unobserved new random variable of interest, f($\chi$;$\beta$) represents fixed effect of known functional form containing unknown parameter vector $\beta$\ulcorner=($\beta$$_1$,…,$\beta$\ulcorner), $\mu$($\chi$) is a random function of mean zero and the known covariance function r(.,.), $\varepsilon$'=($\varepsilon$$_1$,…,$\varepsilon$\ulcorner) is the set of uncorrelated measurement errors with zero mean and unit variance and $\sigma$ is an unknown dispersion(scale) parameter. On the basis of finite-sample, small-dispersion asymptotic framework, we derive an absolute lower bound for the asymptotic mean squared errors of prediction(AMSEP) of the regular-consistent non-linear predictors of the new random variable of interest y*. Then we construct an optimal predictor of y* which attains the lower bound irrespective of types of distributions of random effect $\mu$(.) and measurement errors $\varepsilon$.

• Communications for Statistical Applications and Methods
• /
• 제28권6호
• /
• pp.595-610
• /
• 2021

Recently a few articles have derived robust first-order rotatable and D-optimal designs for the lifetime response having distributions gamma, lognormal, Weibull, exponential assuming errors that are correlated with different correlation structures such as autocorrelated, intra-class, inter-class, tri-diagonal, compound symmetry. Practically, a first-order model is an adequate approximation to the true surface in a small region of the explanatory variables. A second-order model is always appropriate for an unknown region, or if there is any curvature in the system. The current article aims to extend the ideas of these articles for second-order models. Invariant (free of the above four distributions) robust (free of correlation parameter values) second-order rotatable designs have been derived for the intra-class and inter-class correlated error structures. Second-order rotatability conditions have been derived herein assuming the response follows non-normal distribution (any one of the above four distributions) and errors have a general correlated error structure. These conditions are further simplified under intra-class and inter-class correlated error structures, and second-order rotatable designs are developed under these two structures for the response having anyone of the above four distributions. It is derived herein that robust second-order rotatable designs depend on the respective error variance covariance structure but they are independent of the correlation parameter values, as well as the considered four response lifetime distributions.

• 대한수학회보
• /
• 제39권1호
• /
• pp.23-31
• /
• 2002

A scale parameter is the principal parameter to be estimated, since it corresponds to one of the main reliability characteristics, namely the average time to failure. To provide robustness of scale estimators to gross errors in the data, we apply the Huber minimax approach in time to failure models of the statistical reliability theory. The minimax valiance estimator of scale is obtained in the important particular case of the exponential distribution.

• 한국통계학회:학술대회논문집
• /
• 한국통계학회 2002년도 추계 학술발표회 논문집
• /
• pp.79-83
• /
• 2002

A phase III clinical trial of a new drug for neutropenia induced by chemotherapy is presented and consider adding random effects in crossover design which was used in the clinical study. The diagnostics for its heteroscedasticity based on score statistic is derived for detecting homoscedasticity of errors in crossover design. A small simulation study is peformed to investigate the finite sample behaviour of the test statistic which is known to have an asymptotic chi-square distribution under the null hypothesis.

• Communications for Statistical Applications and Methods
• /
• 제8권3호
• /
• pp.739-751
• /
• 2001

In this paper, a sample design is studied for 2000 forestry management survey of five types forestry , tree felling, gathering of pine mushroom, growing of nut trees, growing of wild flowers, and lumbering industry. We introduce population stratification and a modified stratified cut-off sampling which deal with determination of sample size, sample allocation, and estimation of total and variance of estimator. Substitution of sample units and imputation of nonresponse units are discussed for reducing the nonsampling errors.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2001년도 ICCAS
• /
• pp.56.3-56
• /
• 2001

This study describes the jump error smoothing scheme with fuzzy logic based on the scalar adaptive filter. The scalar adaptive filter is an useful algorithm for smoothing abrupt jump errors. However, the performances of scalar adaptive algorithm depend on the variance of real signal. So to design an effective algorithm, many informations of real and jump signal are required. In this paper, the fuzzy rules are designed by the analysis of scalar adaptive filter, and then the improved and simplified scheme is developed for smoothing the jump error. Simulations to INS/GPS integrated system show that the proposed method is effective.