• Nuclear Engineering and Technology
• /
• 제39권3호
• /
• pp.215-220
• /
• 2007

The existing flooding Probabilistic Safety Analysis(PSA) was updated to reflect the Korean plant specific operating experience data into the flooding frequency to improve the PSA quality. Both the Nuclear Power Experience(NPE) database and the Korea Nuclear Pipe Failure Database(NuPIPE) databases were used in this study, and from these databases, only the Pressurized Water Reactor(PWR) data were used for the flooding frequencies of the flooding areas in the primary auxiliary building. With these databases and a Bayesian method, the flooding frequencies for the flooding areas were estimated. Subsequently, the Core Damage Frequency(CDF) for the flooding PSA of the Ulchin(UCN) unit 3 and 4 plants based on the Korean Standard Nuclear Power Plant(KSNP) internal full-power PSA model was recalculated. The evaluation results showed that sixteen flooding events are potentially significant according to the screening criterion, while there were two flooding events exceeding the screening criterion of the existing UCN 3 and 4 flooding PSA. The result was compared with two kinds of cases: (1) the flooding frequency and CDF from the method of the existing flooding PSA with the PWR and Boiled Water Reactor(BWR) data of the NPE database and the Maximum Likelihood Estimate(MLE) method and (2) the flooding frequency and CDF with the NPE database(PWR and BWR data), NuPIPE database, and a Bayesian method. From the comparison, a difference in CDF results was revealed more clearly between the CDF from this study and case (2) than between case (1) and case (2). That is, the number of flooding events exceeding the screen criterion further increased when only the PWR data were used for the primary auxiliary building than when the Korean specific data were used.

• 비파괴검사학회지
• /
• 제18권6호
• /
• pp.464-476
• /
• 1998

원자력 발전소 증기발생기 전열관 검사에 사용되는 와전류 탐상에 있어 결함신호로부터 결함의 형상, 크기, 위치를 정확히 결정하는 것은 매우 중요한 문제 중의 하나이다. 이에 본 연구에서는 유한요소 해석으로 얻은 학습표본으로 훈련시킨 신경회로망을 이용해 이러한 와전류 결함신호의 역문제를 풀었다. 우선 4종류의 축대칭와 전류 결함신호를 총 216개 생성하고, 각각의 결함신호에 대해 24개씩의 와전류 특징을 추출한 후, 그 중에서 결함분석에 유용한 13개의 특징을 선택하였다. 그리고 이렇게 선별된 특징을 기반으로 4가지 형상의 결함에 대한 분류작업을 확률신경 회 로망에 의해 수행하고, 그 결과로 형상이 결정된 결함에 대한 크기산정을 역전파신경 회로망을 사용하여 실시하였다.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2015년도 학술발표회
• /
• pp.225-225
• /
• 2015

Drought events usually evolve slowly in time and their impacts generally span a long period of time. This indicates that the sequence of drought is not completely random. The Hidden Markov Model (HMM) is a probabilistic model used to represent dependences between invisible hidden states which finally result in observations. Drought characteristics are dependent on the underlying generating mechanism, which can be well modelled by the HMM. This study employed a HMM with Gaussian emissions to fit the Standardized Precipitation Index (SPI) series and make multi-step prediction to check the drought characteristics in the future. To estimate the parameters of the HMM, we employed a Bayesian model computed via Markov Chain Monte Carlo (MCMC). Since the true number of hidden states is unknown, we fit the model with varying number of hidden states and used reversible jump to allow for transdimensional moves between models with different numbers of states. We applied the HMM to several stations SPI data in South Korea. The monthly SPI data from January 1973 to December 2012 was divided into two parts, the first 30-year SPI data (January 1973 to December 2002) was used for model calibration and the last 10-year SPI data (January 2003 to December 2012) for model validation. All the SPI data was preprocessed through the wavelet denoising and applied as the visible output in the HMM. Different lead time (T= 1, 3, 6, 12 months) forecasting performances were compared with conventional forecasting techniques (e.g., ANN and ARMA). Based on statistical evaluation performance, the HMM exhibited significant preferable results compared to conventional models with much larger forecasting skill score (about 0.3-0.6) and lower Root Mean Square Error (RMSE) values (about 0.5-0.9).

• 한국물환경학회지
• /
• 제37권5호
• /
• pp.369-380
• /
• 2021

In this work, we analyzed the effects of drought on the water temperature (WT) of Nakdong river basin major river sections using Standardized Precipitation Index (SPI) and WT data. The analysis was carried out on a seasonal basis. After calculating the optimal time scale of the SPI through the correlation between the SPI and WT data, we used the copula theory to model the joint probability distribution between the WT and SPI on the optimal time scale. During spring and fall, the possibility of environmental drought caused by high WT increased in most of the river sections. Notably, in summer, the possibility of environmental drought caused by high WT increased in all river sections. On the other hand, in winter, the possibility of environmental drought caused by low WT increased in most river sections. From the risk map, which quantified the sensitivity of WT to the risk of environmental drought, the river sections Nakbon C, Namgang E, and Nakbon K showed increased stress in the water ecosystem due to high WT when drought occurred in summer. When drought occurred in winter, an increased water ecosystem stress caused by falling WT was observed in the river sections Gilan A, Yongjeon A, Nakbon F, Hwanggang B, Nakbon I, Nakbon J, Nakbon K, Nakbon L, and Nakbon M. The methodology developed in this study will be used in the future to quantify the effects of drought on water quality as well as WT.

• 한국수자원학회논문집
• /
• 제54권9호
• /
• pp.705-718
• /
• 2021

본 연구에서는 위성 원격 감지 자료를 사용하여 생태학적 가뭄의 가능성을 조사하였다. 먼저 MODIS에서 제공하는 정규식생지수와 지표면온도로 부터 식생건강성지수를 추정하였다. 그런 다음 우리나라 전 지역에 분포하고 있는 기상청 ASOS 주요 60개 지점 주변의 산림지역에서 다양한 강수/증발산 시나리오에 따른 식생 관련 가뭄의 가능성을 추정하기 위한 결합 확률모델을 구성하였다. 본 연구의 결과는 대기의 수분공급이 열악하거나 대기의 수분 요구량이 과도한 조건에서 산림 식생과 관련된 가뭄의 위험 패턴을 보여준다. 또한 다양한 기상학적 가뭄 조건에서 산림 식생과 관련된 가뭄 위험의 민감도를 나타낸다. 이러한 연구 결과는 의사 결정권자가 가뭄 위험을 평가하고 온난화 시대에 산림 식생과 관련된 가뭄 완화 전략을 개발할 수 있는 통찰력을 제공할 것으로 기대된다.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2015년도 학술발표회
• /
• pp.173-173
• /
• 2015

The flood water level in tidal river is determined by the joint effects of flood discharge and tidal water levels at downstream boundary. Due to the variable tidal boundary conditions, the evaluated design water levels associated with a certain flood event can be significantly different. To avoid determining of design water levels just by a certain tidal boundary condition and remove the influence of variability in boundary condition from the evaluation of design water levels, a probabilistic approach is considered in this study. This study focuses on the development of a method to evaluate the realistic design water levels in tidal river with taking into account the combined effects of river discharge and tidal level. The flood water levels are described by the joint probability of two driving forces, river discharge and tidal water levels. The developed method is applied to determine design water levels for the tidal reach of the Han River. An unsteady flow model is used to simulate the flow in the reach. To determine design water levels associated with a certain flood event, first, possible boundary conditions are obtained by sampling starting times of tidal level time series; then for each tidal boundary condition, corresponding peak water levels along the channel are computed; and finally, design water levels are determined by computing the expectations of the peak water levels. Two types of tides which are composed by different constituents are assumed (one is composed by $M_2$, and the other one is composed by $M_2$ and $M_2$) at downstream boundary, and two flood events with different maximum flood discharges are considered in this study. It is found that (a) the computed design water levels with two assumed tides have no significant difference for a certain flood event, though variability of peak water levels due to the tidal effect is considerably different; (b) tidal effect can reach to the Jamsil submerged weir and the effect is obvious in the downstream reach of the Singok submerged weir; (c) in the tidally affected reach, the variability of peak water levels due to the tidal effect is greater if the maximum flood discharge is smaller.

• 농업과학연구
• /
• 제48권1호
• /
• pp.33-43
• /
• 2021

Along with the widespread use of pesticides in the world, concerns over human health impacts are rapidly growing. There is a large body of evidence on the relationship between the exposure to pesticides and the elevated rate of chronic diseases such as different types of cancers, diabetes, neurodegenerative disorders like Parkinson, Alzheimer, and amyotrophic lateral sclerosis (ALS), birth defects, and reproductive disorders. This research assessed the health risk of pesticide residues by the dietary intake of vegetables collected from the agro-based markets of Dhaka, Bangladesh. As some of the banned pesticides were also found in vegetable samples, they may pose a higher risk because of cheaper availability and hence the government of Bangladesh should take strong measures to control these banned pesticides. Five organo phosphorus (chlorpyrifos, parathion, ethion, acephate, fenthion) and two carbamate (carbaryl and carbofuran) pesticide residues were identified in twenty four samples of two common vegetables (tomato and brinjal). The pesticide residues ranged from below a detectable limit (< 0.01) to 0.36 mg·kg^-1. Acephate, chlorpyrifos, ethion, and carbaryl were detected in only one sample, while co-occurrence occurred twice for parathion. Continuous monitoring and strict regulation should be enforced regarding the control of pesticide residues in fresh vegetables and other food commodities in Bangladesh.

• Structural Engineering and Mechanics
• /
• 제76권6호
• /
• pp.751-763
• /
• 2020

Available records of recent earthquakes show that near-field earthquakes have different characteristics than far-field earthquakes. In general, most of these unique characteristics of near-fault records can be attributed to their forward directivity. This phenomenon causes the records of ground motion normal to the fault to entail pulses with long periods in the velocity time history. The energy of the earthquake is almost accumulated in these pulses causing large displacements and, accordingly, severe damages in the building. Damage to structures caused by past earthquakes raises the need to assess the chance of future earthquake damage. There are a variety of methods to evaluate building seismic vulnerabilities with different computational cost and accuracy. In the meantime, fragility curves, which defines the possibility of structural damage as a function of ground motion characteristics and design parameters, are more common. These curves express the percentage of probability that the structural response will exceed the allowable performance limit at different seismic intensities. This study aims to obtain the fragility curve for low- and mid-rise structures of reinforced concrete moment frames by incremental dynamic analysis (IDA). These frames were exposed to an ensemble of 18 ground motions (nine records near-faults and nine records far-faults). Finally, after the analysis, their fragility curves are obtained using the limit states provided by HAZUS-MH 2.1. The result shows the near-fault earthquakes can drastically influence the fragility curves of the 6-story building while it has a minimal impact on those of the 3-story building.

• Composites Research
• /
• 제21권6호
• /
• pp.23-30
• /
• 2008

본 논문에서는 자연 노화가 필라멘트 와인딩으로 제작된 압력용기의 강도 분포와 구조 사용 수명에 미치는 영향을 연구하였다. 자연 노화에 따라 변화되는 섬유 방향 파괴 변형률을 설계 확률 변수로 하는 확률 강도 해석을 수행하였다. 이때 확률강도 해석은 정확한 파열 압력을 예측하기 위해 연속 파손 모드가 고려되었고, 비선형 한계식의 해를 구하기 위해 FORM방법이 이용되었다. 해석을 통해 노화 시간별 파괴 확률 분포 선도를 구하였다. 복합재 구조물의 특성상 재료 물성 및 제작 공정 변수 영향으로 제품의 성능 변동성이 비교적 크게 나타났고, 노화로 인한 압력용기의 파열 압력 저하 현상은 대부분 10년 이내에서 발생하였다. 임의 적층의 복합재 압력 용기를 모델로 하여 수명을 평가한 결과, 파괴 확률 2.5%와 안전율 1.3을 고려한 설계 압력 3,250psi기준으로 약 13년의 사용 수명이 평가되었다.

• 한국컴퓨터정보학회논문지
• /
• 제14권2호
• /
• pp.103-110
• /
• 2009

무선 환경이 점점 발전함에 따라 모바일 기기들의 사용이 증가하고 있다. 핸드폰, PDA등 모바일 기기를 이용하여 Cellular망 및 Wibro망 등을 통해 데이터를 주고받고 있다. 그러나 무선통신은 유선통신보다 훨씬 열악한 대역폭을 가지므로, 이러한 한계를 해결하기 위한 방법으로 브로드캐스트 시스템이 대두되고 있다. 보다 효율적인 브로드캐스트 시스템을 구축하기 위해 많은 연구들이 진행되어 왔는데, 그 연구의 한 축은 어떻게 하면 효율적인 브로드캐스트 프로그램을 작성할 것이냐에 대한 문제이다. 이 논문에서는 기존의 요청 확률(Request Probability)을 기반으로 작성된 정적 브로드캐스트 프로그램의 한계를 극복하는 동적 브로드캐스트 프로그램을 제안하고자 한다. 동적 브로드캐스트 프로그램은 실시 간으로 전달되는 클라이언트 요청을 바탕으로 클라이언트의 요청이 반영된 다른 방송을 지속적으로 브로드캐스트하는 시스템을 말한다. 이를 통해 수시로 달라지는 클라이언트의 요구를 지속적으로 반영할 수 있는 브로드캐스트 프로그램을 만들어 보고자 한다. 본 논문에서는 [1]에 발표된 선행연구결과를 확장해 보다 자세한 방법론의 기술과 다양한 각도에서의 성능분석모델을 포함했다.