• 한국지능시스템학회논문지
• /
• 제25권3호
• /
• pp.242-247
• /
• 2015

본 논문은 방사형 기저함수 신경회로망(Radial Basis Function Neural Network) 패턴분류기를 기반으로 강수 에코와 비(非)강수 에코를 분류하는 방법을 제시한다. 강수 에코와 비(非)강수 에코를 분류하기 위하여 기상레이더 자료의 특성을 분석하였다. 이를 기반으로 UF 데이터의 전처리를 실시하여 입력변수(DZ, SDZ, VGZ, SPN, DZ_FR, VR)를 선정 하였고 학습데이터 및 테스트데이터로 구성하였다. 마지막으로, 기상청에서 사용되고 있는 QC 데이터는 제안된 알고리즘의 성능을 비교하기 위해 사용하였다.

• 한국측량학회지
• /
• 제35권5호
• /
• pp.423-430
• /
• 2017

Recently, data analysis research has been carried out using the deep learning technique in various fields such as image interpretation and/or classification. Various types of algorithms are being developed for many applications. In this paper, we propose a precipitation prediction algorithm based on deep learning with high accuracy in order to take care of the possible severe damage caused by climate change. Since the geographical and seasonal characteristics of Korea are clearly distinct, the meteorological factors have repetitive patterns in a time series. Since the LSTM (Long Short-Term Memory) is a powerful algorithm for consecutive data, it was used to predict precipitation in this study. For the numerical test, we calculated the PWV (Precipitable Water Vapor) based on the tropospheric delay of the GNSS (Global Navigation Satellite System) signals, and then applied the deep learning technique to the precipitation prediction. The GNSS data was processed by scientific software with the troposphere model of Saastamoinen and the Niell mapping function. The RMSE (Root Mean Squared Error) of the precipitation prediction based on LSTM performs better than that of ANN (Artificial Neural Network). By adding GNSS-based PWV as a feature, the over-fitting that is a latent problem of deep learning was prevented considerably as discussed in this study.

• 한국농림기상학회지
• /
• 제4권4호
• /
• pp.237-247
• /
• 2002

Air pollution by acid pollutants is problematic in the whole world. Water acidification has already been deteriorating the forest ecosystem. This study was conducted to analyze the acidity and chemical composition of the open precipitation and throughfall at forests with various geographic locations in Korea. The results of this study are as follows; The open precipitation pH was lowest in Seochun. The throughfall pH showed some buffering capacity in only Quercus mongolica stands. In Pinus rigida(Shingal and Seochun) stands, there was little difference from the open precipitation. Chemical composition of the open precipitation for each sampling site showed that $Ca^{2+}$, N $H_{4}$$^{+}$ and S $O_{4}$$^{2-}$ concentrations had higher value than other ions, and except these ions, the small quantity of ions showed different properties to each site. Changes of ion concentrations in the throughfall showed a tendency to increase. ion concentrations of the throughfall increased with washout and nutrient leaching from the trees. In conclusion, the influence was extended to the pure zone, and the frequency of acid rain is increasing. But, if the deposition of pollutants exceeds the capacity of purification, it would damage forest ecosystem. Further investigations are necessary to identify tolerant tree species to acid pollutants.nts.

• 대한토목학회논문집
• /
• 제34권4호
• /
• pp.1139-1150
• /
• 2014

본 연구에서는 M-PRISM 모형을 이용하여 $1km{\times}1km$ 공간해상도 일강수량 추정에 대한 적용성을 검토하였다. 또한 회귀모형을 이용하여 M-PRISM 모형 매개변수를 추정하였으며, 잭나이프 방법을 이용하여 모형을 검증하였다. 기상청 385개 강수 관측지점에 대하여 M-PRISM을 이용하여 일강수량을 추정하고 PRISM 모형과 비교하였다. 비교결과, 강수의 정량적 크기를 추정에서는 두 모형에서 뚜렷한 차이를 찾아볼 수 없었으나, 강수의 발생빈도 추정에 있어서는 M-PRISM 모형이 더 우수한 결과를 나타내었다. 따라서 본 연구에서 제안한 M-PRISM 모형은 고해상도의 일강수량을 추정함에 있어서 매우 유용하게 사용될 수 있을 것으로 판단된다.

• 한국재료학회지
• /
• 제29권6호
• /
• pp.343-348
• /
• 2019

Structural and mechanical effects of the dynamical precipitation in two copper-base alloys have been investigated over a wide range of deformation temperatures. Basing upon the information gained during the experiment, also some general conclusion may be formulated. A one concerns the nature of dynamic precipitation(DP). Under this term it is commonly understood decomposition of a supersaturated solid solution during plastic straining. The process may, however, proceed in two different ways. It may be a homogeneous one from the point of view of distribution and morphological aspect of particles or it may lead to substantial difference in shape, size and particles distribution. The effect is controlled by the mode of deformation. Hence it seems to be reasonable to distinguish DP during homogeneous deformation from that which takes place in heterogeneously deformed alloy. In the first case the process can be analyzed solely in terms of particle-dislocation-particle interrelation. Much more complex problem we are facing in heterogeneously deforming alloy. Deformation bands and specific arrangement of dislocations in form of pile-ups at grain boundaries generate additional driving force and additional nucleation sites for precipitation. Along with heterogeneous precipitation, there is a homogeneous precipitation in areas between bands of coarse slip which also deform but at much smaller rate. This form of decomposition is responsible for a specially high hardening rate during high temperature straining and for thermally stable product of the decomposition of alloy.

• 물과 미래
• /
• 제27권3호
• /
• pp.71-82
• /
• 1994

수자원 계획을 위하여 기본적으로 필요한 과거 강수량자료의 통계적 특성, 역년(calendar year)과 수문년(water year)의 연강수량 관계, 기간별 총강수량의 빈도 등을 장기간 과거 강수량을 이용하여 분석하였다. 또한 우리나라의 수자원 계획에 많이 이용해 왔던 1967-1968년 한발기간의 강수량을 분석하였다. 대상유역은 한강, 낙동강, 금강, 섬진강, 영산강 유역으로, 기상청 65개 우량관측소의 1905-1968년 한발기간의 강수량을 분석하였다. 대상유역은 한강, 낙동강, 금강, 섬진강, 영산강 유역으로, 기상청 65개 우량관측소의 1905-1991년 기간 자료를 이용하였으며, Thiessen 가중법으로 유역평균강수량을 산정하여 분석하였다. 본 연구의 결과에서 우리나라의 연강 수량은 전체적으로 증가하는 경향이 있었으나 통계적 검정결과 그 변동량의 유의성이 없었다. 역년과 수문년의 연강수량 관계식을 제시하였으며, 두 기간의 연강수량은 거의 차이가 없는 것으로 나타났다. 3, 6, 9 그리고 12개월 기간에 따른 총강수량의 연 최저치계열을 작성하였고, 2변수 대수정규분포를 이용하여 각 기간별 빈도강수량을 제시하였다. 1967-1968년 강우분석의 기준으로 볼 때, 댐 등에 의한 수자원 개발이 않된 자연하천 유역에서 건기(10-5월) 또는 우기(6-9월)의 총강수량이 과거 평균수량의 약 75%정도를 기록하면 한발를 초래하고, 약 60% 정도의 강수량이면 심한 한발을 초래한다고 할 수 있다.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2009년도 학술발표회 초록집
• /
• pp.112-115
• /
• 2009

The research of climate change impact in hydrometeorology often relies on climate change information. In this paper, neural networks models such as support vector machine neural networks model (SVM-NNM) and multilayer perceptron neural networks model (MLP-NNM) are proposed statistical downscaling of the monthly precipitation. The input nodes of neural networks models consist of the atmospheric meteorology and the atmospheric pressure data for 2 grid points including $127.5^{\circ}E/35^{\circ}N$ and $125^{\circ}E/35^{\circ}N$, which produced the best results from the previous study. The output node of neural networks models consist of the monthly precipitation data for Seoul station. For the performances of the neural networks models, they are composed of training and test performances, respectively. From this research, we evaluate the impact of SVM-NNM and MLP-NNM performances for the downscaling of the monthly precipitation data. We should, therefore, construct the credible monthly precipitation data for Seoul station using statistical downscaling method. The proposed methods can be applied to future climate prediction/projection using the various climate change scenarios such as GCMs and RCMs.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2009년도 학술발표회 초록집
• /
• pp.125-128
• /
• 2009

The research of climate change impact in hydrometeorology often relies on climate change information. In this paper, neural networks models such as generalized regression neural networks model (GRNNM) and multilayer perceptron neural networks model (MLP-NNM) are proposed statistical downscaling of the daily precipitation. The input nodes of neural networks models consist of the atmospheric meteorology and the atmospheric pressure data for 4 grid points including $127.5^{\circ}E/37.5^{\circ}N$, $127.5^{\circ}E/35^{\circ}N$, $125^{\circ}E/37.5^{\circ}N$ and $125^{\circ}E/35^{\circ}N$, respectively. The output node of neural networks models consist of the daily precipitation data for Seoul station. For the performances of the neural networks models, they are composed of training and test performances, respectively. From this research, we evaluate the impact of GRNNM and MLP-NNM performances for the downscaling of the daily precipitation data. We should, therefore, construct the credible daily precipitation data for Seoul station using statistical downscaling method. The proposed methods can be applied to future climate prediction/projection using the various climate change scenarios such as GCMs and RCMs.

• 한국수자원학회논문집
• /
• 제38권10호
• /
• pp.885-894
• /
• 2005

1970년대 이후, 우리나라는 산업화에 따른 급격한 도시화가 이루어졌다. 본 논문에서는 우리나라의 대표적인 도시인 서울특별시 및 6대 광역시의 1973년부터 2003년까지의 31개년의 강수랑 자료를 이용하여 강수량의 변화에 대하여 분석하였다. 이와 함께 도시화에 따른 강수량의 변동성을 평가하기 위해서 비도시 지역을 선정하였으며 도시 지역의 강수량 변화와 비교하였다. 도시 지역과 비도시 지역의 연강수량, 계절별 강수량, 지속 시간 1시간 및 24시간연최대 강수량에 대해 임의기간에 따른 평균 분석, 경향성 분석, 변동성 분석, 비매개변수적 빈도 해석을 수행한 결과, 도시화 지역에서 비도시화 지역보다 강우 증가율이 더 컸으며, 특히 여름 강수량의 증가량이 두드러졌다.

• Water Engineering Research
• /
• 제5권3호
• /
• pp.111-121
• /
• 2004

For the long-term strategic water resources planning, forecasting the future streamflow change is important to meet the demand of a growing society. The streamflow variation to the decade-long precipitation was investigated for the two major stage gauging stations in Korea. Precipitation and runoff characteristics have been analyzed at Yongwol stream stage in the Han River as well as Sutong stream stage in the Kum River for the future water resources management strategies. Monte Carlo method has been applied to estimate the future precipitation and runoff. Based on the trend line of 10-year moving average of runoff depth for the historical runoff records, the relation between runoff and the time variation was examined in more detail using regression analysis. This study showed that the surface flows have been significantly decreased while precipitation has been stable in these basins. Decreasing in runoff reflects the regional watershed characteristics such as forest cover changes. The findings of this study could contribute to the planning and development for the efficient water resources utilization.