• Ocean and Polar Research
• /
• 제43권1호
• /
• pp.1-14
• /
• 2021

In this study, extreme wind speeds in the Western North Pacific (WNP) were estimated using reanalysis wind fields synthesized with an empirical typhoon vortex model. Reanalysis wind data used is the Fifth-generation European Centre for Medium-Range Weather Forecasts (ECMWF) reanalysis (ERA5) data, which was deemed to be the most suitable for extreme value analysis in this study. The empirical typhoon vortex model used has the advantage of being able to realistically reproduce the asymmetric winds of a typhoon by using the gale/storm-forced wind radii information in the 4 quadrants of a typhoon. Using a total of 39 years of the synthesized reanalysis wind fields in the WNP, extreme value analysis is applied to the General Pareto Distribution (GPD) model based on the Peak-Over-Threshold (POT) method, which can be used effectively in case of insufficient data. The results showed that the extreme analysis using the synthesized wind data significantly improved the tendency to underestimate the extreme wind speeds compared to using only reanalysis wind data. Considering the difficulty of obtaining long-term observational wind data at sea, the result of the synthesized wind field and extreme value analysis developed in this study can be used as basic data for the design of offshore structures.

• 문화기술의 융합
• /
• 제8권1호
• /
• pp.531-536
• /
• 2022

본 논문은 꼬리가 두꺼운 분포의 꼬리부분에 대한 분포를 추정할 경우 모수적 방법과 비모수적 방법의 성능에 대해 비교하였다. 모수적 방법으로는 일반화 극단값 분포와 일반화 파레토 분포를 이용하였고, 비모수적 방법은 커널형 확률밀도함수 추정방법을 적용하였다. 두 접근법의 비교를 위해 2014년부터 2018년까지 서울시 관측소별 일일 미세먼지 공공데이터를 이용하여 블록 최댓값 모형과 분계점 초과치 모형을 적용하여 함수 추정한 결과를 함께 보이고 2년, 5년, 10년의 재현수준을 통해 고농도의 미세먼지가 일어날 지역을 예측하였다.

• 산업기술연구
• /
• 제24권A호
• /
• pp.215-226
• /
• 2004

A dynamic reliability model which can take into account the time history of loading sequences may be applied to the analyses of the hydraulic stability of armor units on rubble-mound breakwaters. All the parameters related to the stability of structures have been considered to be constants in the deterministic model until now. Thus, it is impossible to study the effects of some uncertainties of the related random variables on the stability of structures. In this paper, the dynamic reliability model can be developed by POT(Peak Over Threshold) method in order to take into account the time history of loading sequences and to investigate the temporal behaviors of stability of structure with its loading history. Finally, it is confirmed that the results of dynamic reliability model agree with straight- forwardly those of AFDA(Approximate Full Distribution Approach) of the static reliability model for the same input conditions. In addition, the temporal behaviors of probability of failure can be studied by the dynamic reliability model developed to analyze the hydraulic stability of armor units on rubble-mound breakwaters. Therefore, the present results may be useful for the management of repair and maintenance over the whole life cycle of structure.

• 재무관리연구
• /
• 제22권1호
• /
• pp.119-146
• /
• 2005

본 연구는 극한치 이론을 적용하여 국내 주식시장에 대한 VAR값을 구하고 이의 유용성을 살펴보았다. 극한치모형으로는 블록최대값모형과 POT 모형을 이용하였고 백테스트를 통하여 이들 모형의 적정성을 알아보았다. 극한치모형 중 블록최대값 모형은 신뢰수준의 변화에 따라 VAR 추정치의 변동이 매우 큰 것으로 나타났으며, 블록의 크기를 어떻게 선택하는가에 따라 모수의 추정치가 크게 달라져 VAR값의 안정성이 떨어지는 것으로 나타났다. 이는 국내 주식시장에 대해 VAR 측정시 블록최대값 모형을 사용하는 것은 적절치 않음을 시사하는 것이다. 반면 POT모형은 임계치의 선택에 따라서 VAR 값이 달라지기는 하나 상대적으로 안정적인 모습을 보이며, 또한 백테스트 결과 97.5% 이상의 신뢰수준에서 VAR값이 델타 VAR에 비하여 우수한 성과를 보이는 것으로 나타났다. 특히 POT모형은 신뢰수준이 높아질수록 그 우월성이 높은 것으로 나타나, 주로 99% 이상의 높은 신뢰수준의 VAR값을 이용하는 경우에 위험의 관리수단으로 유용한 모형임을 시사하고 있다. 또한 극한치모형은 수익률의 왼쪽 꼬리와 오른쪽 꼬리를 분리하여 추정하고 이를 VAR의 계산에 이용하기 때문에, 수익률분포가 비대칭적 특징을 보이는 우리나라 주식시장의 VAR 측정시 적절한 방법임을 확인할 수 있었다.

• 한국환경과학회지
• /
• 제10권3호
• /
• pp.217-223
• /
• 2001

Where no records are available at a site, a preliminary estimate may be made from relations between floods and catchment chatacteristics. A number of these chatacteristics were chosen for testing and were measured for those catchments where mean annual flood estimates were available. Although the improvement using extended data in regression of flood estimates on catchment characteristics was small, this may be due to the limitations of the regression model. When an individual short term record is to be extended, more detailed attention can be given; an example is presented of the technique which should be adopted in practice, particularly when a short term record covers a period which is known to be biassed. A method of extending the peaks over a threshold series is presented with a numerical example. The extension of records directly from rainfall by means of a conceptual model is discussed, although the application of such methods is likely to be limited by lack of recording raingauge information. Methods of combining information from various sources are discussed in terms of information from catchment characteristics supplemented by records. but are generally applicable to different sources of information. The application of this technique to estimating the probable maximum flood requires more conservative assumptions about the antecedent condition, storm profile and unit hydrograph. It is suggested that the profile and catchment wetness index at the start of the design duration should be based on the assumption that the estimated maximum rainfall occurs in all durations centered on the storm peak.

• 한국수자원학회논문집
• /
• 제45권3호
• /
• pp.275-290
• /
• 2012

본 연구에서는 극치강우의 시간분포 연구를 위하여 서울지점 우량관측소의 자기기록지를 1분단위로 독취한 MMR(minutely data using the magnetic recording)자료와 최근 들어 관측을 시작한 AWS (automatic weather system) 분단위기상관측 자료를 이용하여 연최대치 계열의 중앙값을 기준으로 한 POT(peaks over threshold) 계열 추출을 통하여 강우의 최적 시간분포 모형을 개발하였다. 기존 Huff 방법에서의 최대 단점인 지속기간별 시간분포 변화 특성을 고려하지 못하는 점과 강우사상별 강우총량에 대한 기준강우량의 일괄적용 등의 문제를 개선하였으며, 분단위 관측자료의 가중치 적용을 통한 순위결정으로 최빈분위를 선택하고 IQR (interquartile range) matrix의 적용을 통한 Quartile별 호우사상을 추출하는 방법을 제안하였다. 마지막으로 추출된 분단위 무차원 단위우량주상도에 핵밀도함수를 적용하여 자료의 크기와 분포 특성을 고려한 지속기간별 최적 시간분포형을 유도하였다.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2008년도 International Symposium on Remote Sensing
• /
• pp.430-432
• /
• 2008

Visible channel calibration method using deep convective clouds (DCCs) is developed. The method has advantages that visible radiance is not sensitive to cloud optical thickness (COT) for deep convective clouds because visible radiance no longer increases when COT exceeds 100. Therefore, once DCCs are chosen appropriately, and then cloud optical properties can be assumed without operational ancillary data for the specification of cloud conditions in radiative transfer model. In this study, it is investigated whether IR measurements can be used for the selection of DCC targets. To construct appropriate threshold value for the selection of DCCs, the statistics of cloud optical properties are collected with MODIS measurements. When MODIS brightness temperature (TB) at 11 ${\mu}$ m is restricted to be less than 190 K, it is shown that more than 85% of selected pixels show COT ${\geq}$ 100. Moreover, effective radius ($r_e$) distribution shows a sharp peak around 20 ${\mu}m$. Based on those MODIS observations, cloud optical properties are assumed as COT = 200 and $r_e$ = 20 ${\mu}m$ for the simulation of MODIS visible (0.646 ${\mu}m$) band radiances over DCC targets.

• Wind and Structures
• /
• 제19권4호
• /
• pp.421-441
• /
• 2014

With the assistance of typhoon field data at aerial elevation level observed by meteorological satellites and wind velocity and direction records nearby the ground gathered in Guangzhou Weather Station between 1985 and 2001, some key wind field parameters under typhoon climate in Guangzhou region were calibrated based on Monte-Carlo stochastic algorithm and Meng's typhoon numerical model. By using Peak Over Threshold method (POT) and Generalized Pareto Distribution (GPD), Wind field characteristics during typhoons for various return periods in several typical engineering fields were predicted, showing that some distribution rules in relation to gradient height of atmosphere boundary layer, power-law component of wind profile, gust factor and extreme wind velocity at 1-3s time interval are obviously different from corresponding items in Chinese wind load Codes. In order to evaluate the influence of typhoon field parameters on long-span flexible bridges, 1:100 reduced-scale wind field of type B terrain was reillustrated under typhoon and normal conditions utilizing passive turbulence generators in TJ-3 wind tunnel, and wind-induced performance tests of aero-elastic model of long-span Guangzhou Xinguang arch bridge were carried out as well. Furthermore, aerodynamic admittance function about lattice cross section in mid-span arch lib under the condition of higher turbulence intensity of typhoon field was identified via using high-frequency force-measured balance. Based on identified aerodynamic admittance expressions, Wind-induced stochastic vibration of Xinguang arch bridge under typhoon and normal climates was calculated and compared, considering structural geometrical non-linearity, stochastic wind attack angle effects, etc. Thus, the aerodynamic response characteristics under typhoon and normal conditions can be illustrated and checked, which are of satisfactory response results for different oncoming wind velocities with resemblance to those wind tunnel testing data under the two types of climate modes.

• Nuclear Engineering and Technology
• /
• 제46권6호
• /
• pp.837-846
• /
• 2014

A lead slowing down spectrometer (LSDS) is under development for analysis of isotopic fissile material contents in pyro-processed material, or spent fuel. Many current commercial fissile assay technologies have a limitation in accurate and direct assay of fissile content. However, LSDS is very sensitive in distinguishing fissile fission signals from each isotope. A neutron spectrum analysis was conducted in the spectrometer and the energy resolution was investigated from 0.1eV to 100keV. The spectrum was well shaped in the slowing down energy. The resolution was enough to obtain each fissile from 0.2eV to 1keV. The detector existence in the lead will disturb the source neutron spectrum. It causes a change in resolution and peak amplitude. The intense source neutron production was designed for ~E12 n's/sec to overcome spent fuel background. The detection sensitivity of U238 and Th232 fission chamber was investigated. The first and second layer detectors increase detection efficiency. Thorium also has a threshold property to detect the fast fission neutrons from fissile fission. However, the detection of Th232 is about 76% of that of U238. A linear detection model was set up over the slowing down neutron energy to obtain each fissile material content. The isotopic fissile assay using LSDS is applicable for the optimum design of spent fuel storage to maximize burnup credit and quality assurance of the recycled nuclear material for safety and economics. LSDS technology will contribute to the transparency and credibility of pyro-process using spent fuel, as internationally demanded.

• 한국해안·해양공학회논문집
• /
• 제31권3호
• /
• pp.115-128
• /
• 2019

본 논문에서는 공정별 해상작업 가능 기간의 합리적 산출이 가능한 확률모형이 제시된다. 확률모형을 유도하기 위해, 먼저 JMA(Japan Meterological Agency)와 NOAA(National Oceanic and Atmospheric Administration)의 해상풍 자료와 SWAN에 기초하여 2003년 1월 1일부터 2017년 12월 31일까지 한 시간 간격으로 울산 전면 해역에서의 유의 파고와 첨두 주기를 역추산 하였다. 이어 모의된 유의파고 시계열 자료로부터 최소 자승법을 활용하여 장기 유의파고 확률분포를 도출하였으며, 해석결과 그 동안 선호되던 삼 변량 Weibull 분포보다는 수정 Glukhovskiy 분포 계열에서 일치도가 가장 우월하였다. 보다 정확한 확률모형의 개발 가능성을 검토하기 위해 Borgman 선회적분을 활용하여 역 추산 단위 간격인 한 시간 내에서 출현하는 개별 파랑이 고려된 파고분포도 함께 유도하였다. 수정 Glukhovskiy 분포의 모수는 $A_p=15.92$, $H_p=4.374m$, ${\kappa}_p=1.824$로 드러났으며 해상작업 한계 파고가 $H_S=1.5m$인 경우 작업가능일 수는 319일로 모의되었다. 이와 더불어 확률모형의 검증자료를 얻기 위해 파고가 해상 준설작업 한계 파고로 기 보고된 바 있는 $H_S=1.5m$(Lee, 1991)를 상회하여 지속되는 시간을 유의파고 시계열 자료를 파별분석(wave by wave analysis)하여 산출하였다. 산출결과 2003년부터 2017년까지의 평균 지속기간은 45.5일로 확률모형으로부터 산출된 기간에 상당히 근접하였다.