• Proceedings of the Korea Water Resources Association Conference
• /
• 2015.05a
• /
• pp.501-501
• /
• 2015

댐과 같은 수공구조물의 치수능력부족은 구조물의 파손이나 붕괴로 직결되며, 대규모 재산피해와 인명피해가 불가피하다. 또한 최근 기후변화 현상에 의해 발생되고 있는 예상치 못한 큰 강우와 홍수는 댐 안전을 저하하는 요소로 간주되며, 복합적으로 발생시 댐의 치수능력이 크게 저하되어 댐 파괴에 영향을 미칠 가능성이 크다. 또한 ICOLD, 세계대댐회 등의 조사 결과 댐 파괴에 큰 영향을 미치고 있는 수문학적 요소로서 예상치 못한 강우, 홍수 및 월류로 조사되었다. 이러한 이유로 수문학적 위험인자를 효과적으로 고려하기 위해 2000년도 이후 선진국에서는 위험도 해석 기법을 기반으로 파괴모드, 다양한 하중조건 등을 조합하여 위험도 해석을 통해 댐의 안전도 검토를 실시하고 있다. 따라서 최근 증가하는 기상변동성을 능동적으로 고려하기 위해서는 위험도 해석기반의 수공구조물 안정성 평가기법을 기반으로 하는 종합적인 위험도 해석 방안수립이 요구된다. 이러한 이유로 본 연구에서는 BN 모형 기반의 댐 위험도 적용에 앞서 실증댐을 대상으로 분석을 수행하였으며, 분석 절차는 다음과 같다. 첫째, ICOLD 및 세계대댐회 등 다양한 논문, 보고서 등을 조사하여 댐 붕괴에 가장 큰 영향을 미치는 수문학적 파괴인자를 도출하였다. 둘째, BN 모형 구축시 각 노드는 앞서 도출된 수문학적 파괴인자를 이용하였으며, 각 파괴인자에 적합한 확률분포형을 부여하였다. 마지막으로, 노드별 임계값을 부여하여 상황별 파괴인자의 변화 양상을 도출하였다. 본 연구의 결과로 인해 연구대상댐의 가장 취약한 수문학적 파괴인자 도출이 가능하며, 댐의 보수 보강시 우선순위 의사결정의 지원이 가능 할 것으로 판단된다.

• The Korean Journal of Applied Statistics
• /
• v.34 no.3
• /
• pp.491-505
• /
• 2021

We consider linear regression models in high-dimensional settings (p ≫ n) and compare various classes of priors. The spike and slab prior is one of the most widely used priors for Bayesian regression models, but its model space is vast, resulting in a bad performance in finite samples. As an alternative, various continuous shrinkage priors, including the horseshoe prior and its variants, have been proposed. Although each of the above priors has been investigated separately, exhaustive comparative studies of their performance have been conducted very rarely. In this study, we compare the spike and slab prior, the horseshoe prior and its variants in various simulation settings. The performance of each method is demonstrated in terms of the regression coefficient estimation and variable selection. Finally, some remarks and suggestions are given based on comprehensive simulation studies.

• Journal of Korea Water Resources Association
• /
• v.52 no.9
• /
• pp.589-601
• /
• 2019

Various applications of radar rainfall data have been actively employed in the field of hydro-meteorology. Since radar rainfall is estimated by using predefined reflectivity-rainfall intensity relationships, they may not have sufficient reproducibility of observations. In this study, a generalized linear model is introduced to better capture the Z-R relationship in the context of bias correction within a Bayesian regression framework. The bias-corrected radar rainfall with the generalized linear model is more accurate than the widely used mean field bias correction method. In addition, we analyzed variability of the bias correction parameters under various geomorphological conditions such as the height of the weather station and the separation distance from the radar. The identified relationship is finally used to derive a regionalized formula which can provide bias correction factors over the entire watershed. It can be concluded that the bias correction parameters and regionalized method obtained from this study could be useful in the field of radar hydrology.

• The Korean Journal of Applied Statistics
• /
• v.11 no.2
• /
• pp.377-387
• /
• 1998

A Markov Chain Monte Carlo method with data augmentation is developed to compute the features of the posterior distribution. For each observed failure epoch, we introduced latent variables that indicates with component of the Superposition model. This data augmentation approach facilitates specification of the transitional measure in the Markov Chain. Metropolis algorithms along with Gibbs steps are proposed to preform the Bayesian inference of such models. for model determination, we explored the Pre-quential conditional predictive Ordinate(PCPO) criterion that selects the best model with the largest posterior likelihood among models using all possible subsets of the component intensity functions. To relax the monotonic intensity function assumptions, we consider in this paper Superposition of Musa-Okumoto and Erlang(2) models. A numerical example with simulated dataset is given.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2019.05a
• /
• pp.57-57
• /
• 2019

최근 수문기상학 분야에서 레이더 강수량을 활용한 응용연구가 활발하게 진행되고 있다. 하지만 레이더 강수량은 경험적으로 설정된 레이더 반사도-강우강도 관계식을 활용하여 레이더 강수량을 산정하기 때문에 실제지상에 도달하는 강수량과 정량적인 오차가 필연적으로 발생한다. 따라서 고해상도의 레이더 강수량을 활용한 신뢰도 높은 수문해석을 위하여 레이더 강수량의 편의보정이 필수적으로 선행되어야한다. 본 연구에서는 불확실성을 고려한 레이더 강수량 편의보정을 위하여 Bayesian 추론기법과 일반화 선형모형(generalized linear model)을 연계하여 레이더 강수량 편의보정 매개변수를 산정하였다. 일반화 선형모형을 적용한 레이더 강수량 편의보정 결과는 현재 널리 사용되고 있는 평균보정(mean field bias) 기법에 비하여 통계지표가 개선된 레이더 강수량 편의보정 결과를 도출하였다. 추가적으로 지형학적 특성에 따른 레이더 강수량 편의보정 매개변수의 변동성을 분석하여 고도 및 이격거리에 따른 편의보정 매개변수의 지역화 공식을 제시하였다. 본 연구를 통하여 개발된 레이더 강수량 편의보정 매개변수 산정 및 지역화 연구는 레이더 관측전략 수립과정에 유용한 기초자료로 활용될 것으로 판단된다.

• Journal of Korea Water Resources Association
• /
• v.51 no.9
• /
• pp.769-782
• /
• 2018

Estimation of design floods is typically required for hydrologic design purpose. Design floods are routinely estimated for water resources planning, safety and risk of the existing water-related structures. However, the hydrologic data, especially streamflow data for the design purposes in South Korea are still very limited, and additionally the length of streamflow data is relatively short compared to the rainfall data. Therefore, this study collected a large number design flood data and watershed characteristics (e.g. area, slope and altitude) from the national river database. We further explored to formulate a scaling approach for the estimation of design flood, which is a function of the watershed characteristics. Then, this study adopted a Hierarchical Bayesian model for evaluating both parameters and their uncertainties in the regionalization approach, which models the hydrologic response of ungauged basins using regression relationships between watershed structure and model. The proposed modeling framework was validated through ungauged watersheds. The proposed approach have better performance in terms of correlation coefficient than the existing approach which is solely based on area as a predictor. Moreover, the proposed approach can provide uncertainty associated with the model parameters to better characterize design floods at ungauged watersheds.

• Journal of the Korean Geographical Society
• /
• v.49 no.5
• /
• pp.761-778
• /
• 2014

There has been relatively little study to model price for commercial property because of its low transaction volume in the market. Despite of this thin market character, this paper tried to estimate prices for commercial lots as accurate as possible. We constructed a model whose components consist of mean structure(global trend), exponential covariance function and a pure error term, and applied it to actual sales price data of Seoul. We explicitly took account of spatial autocorrelation of land price by utilizing a kriging technique, a representative method of spatial interpolation, because the land price of commercial lots has feature of differential price forming pattern depending on submarkets they belong to. In addition, we chose to apply a bayesian kriging to overcome data scarcity by incorporating experts' knowledge into prior probability distribution. The chosen model's excellent performance was verified by the result from validation data. We confirmed that the excellence of the model is attributed to incorporating both autocorexperts' knowledge and spatial autocorrelation in the model construction. This paper is differentiated from previous studies in the sense that it applied the bayesian kriging technique to estimate price for commercial lots and explicitly combined experts' knowledge with data. It is expected that the result of this paper would provide a useful guide for the circumstances under which property price has to be estimated reliably based on sparse transaction data.

• Korean Journal of Remote Sensing
• /
• v.15 no.2
• /
• pp.147-158
• /
• 1999

Remote sensing technique has offered better understanding of our environment for the decades by providing useful level of information on the landcover. In many applications using the remotely sensed data, digital image processing methodology has been usefully employed to characterize the features in the data and develop the models. Random field models, especially Markov Random Field (MRF) models exploiting spatial relationships, are successfully utilized in many problems such as texture modeling, region labeling and so on. Usually, remotely sensed imagery are very large in nature and the data increase greatly in the problem requiring temporal data over time period. The time required to process increasing larger images is not linear. In this study, the methodology to reduce the computational cost is investigated in the utilization of the Markov Random Field. For this, multiresolution framework is explored which provides convenient and efficient structures for the transition between the local and global features. The computational requirements for parameter estimation of the MRF model also become excessive as image size increases. A Bayesian approach is investigated as an alternative estimation method to reduce the computational burden in estimation of the parameters of large images.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2011.05a
• /
• pp.448-452
• /
• 2011

기존의 하천개수사업 치수경제성분석에서는 홍수피해경감편익 산정시 계획홍수위 이하의 홍수에 대해서 제방이 완벽히 방어한다는 가정 하에 제방으로 인한 피해경감액을 편익으로 산정하고 있다. 그러나 전통적 빈도해석 방법 및 수리수문 모형에 내재된 매개변수 불확실성으로 인하여 특정 하천구간에서 산정된 계획빈도 이하의 홍수위가 제방고에 해당하는 임계사상을 초과할 수도, 반대로 계획빈도 이상의 홍수위가 임계사상을 초과하지 않을 가능성도 있다. 이러한 불확실성은 수공구조물의 붕괴에 대한 잠재성을 가진 중요한 요인으로도 작용한다. 본 연구는 이러한 잠재적 위험도를 제방 월류위험도로 정의하고 이를 Bayesian MCMC에 의해 산정하는 방법을 제시하였다. 제시된 방법론은 4대강살리기사업 전 후에 대해 적용하였으며, 계획홍수위 저하에 따른 잠재적 홍수위험 감소 효과를 정량적으로 나타낼 수 있었다. 월류 위험도는 빈도별 홍수피해액의 피해발생 확률로서도 적용될 수 있으며, 이는 물리적 침수구역 설정의 어려움에 따른 홍수피해액 과다산정 문제 해결의 대안으로서도 의미가 있다.

• Journal of Environmental Science International
• /
• v.27 no.7
• /
• pp.499-507
• /
• 2018

Agricultural meteorological information is an important resource that affects farmers' income, food security, and agricultural conditions. Thus, such data are used in various fields that are responsible for planning, enforcing, and evaluating agricultural policies. The meteorological information obtained from automatic weather observation systems operated by rural development agencies contains missing values owing to temporary mechanical or communication deficiencies. It is known that missing values lead to reduction in the reliability and validity of the model. In this study, the hierarchical Bayesian spatio-temporal model suggests replacements for missing values because the meteorological information includes spatio-temporal correlation. The prior distribution is very important in the Bayesian approach. However, we found a problem where the spatial decay parameter was not converged through the trace plot. A suitable spatial decay parameter, estimated on the bias of root-mean-square error (RMSE), which was determined to be the difference between the predicted and observed values. The latitude, longitude, and altitude were considered as covariates. The estimated spatial decay parameters were 0.041 and 0.039, for the spatio-temporal model with latitude and longitude and for latitude, longitude, and altitude, respectively. The posterior distributions were stable after the spatial decay parameter was fixed. root mean square error (RMSE), mean absolute error (MAE), mean absolute percentage error (MAPE), and bias were calculated for model validation. Finally, the missing values were generated using the independent Gaussian process model.