• 대한교통학회지
• /
• 제5권1호
• /
• pp.59-74
• /
• 1987

A prediction is an indispensable element to research of Social Science, especially in Regional planning, City planning, and Transportation planning. Since 1930s, varieties of prediction methods have been developed. In the 1980s, numerical models have been used by high-developed computers. even though the numerical models can be figured mathematically, it could not be applied practically due to it's expertness and complicateness. And even professional planners often can not use their ideas which are valuable experiences in prediction process, because they are not knowledgable for numerical models. The YSIM developed by author, is available as follows. i)Numerical modeling of professional experiences ii)Providing a foundation of large-scale model iii) Understanding of research object structure The YSIM make use of matrix to identify the system structure which is similar to the Cross Impact Method. To evaluated the YSIM availabilities, it is compared with the early developed methodologies such as KSIM, QSIM, and SPIN. As the result, it was confirmed that YSIM was more accurate in the prediction. The algorithms in YSIM is programmed for use of PCs.

• Wind and Structures
• /
• 제18권6호
• /
• pp.619-631
• /
• 2014

This paper proposes new criteria to fix hidden neuron in Multilayer Perceptron Networks for wind speed prediction in renewable energy systems. To fix hidden neurons, 101 various criteria are examined based on the estimated mean squared error. The results show that proposed approach performs better in terms of testing mean squared errors. The convergence analysis is performed for the various proposed criteria. Mean squared error is used as an indicator for fixing neuron in hidden layer. The proposed criteria find solution to fix hidden neuron in neural networks. This approach is effective, accurate with minimal error than other approaches. The significance of increasing the number of hidden neurons in multilayer perceptron network is also analyzed using these criteria. To verify the effectiveness of the proposed method, simulations were conducted on real time wind data. Simulations infer that with minimum mean squared error the proposed approach can be used for wind speed prediction in renewable energy systems.

• 한국물환경학회지
• /
• 제32권6호
• /
• pp.552-561
• /
• 2016

For the prediction of hydrologic phenomenon, predicting future land use change is a very important task. This study aimed to compare and analyze the two land use change models, CLUE-S and SLEUTH3-R. The analysis of two models were performed based on the MSR value such that the model with more reliable MSR value can be recommended as an appropriate land use change prediction model. The model performance was examined by applying to the Gapcheon A watershed. Land use map of the study area of 2007 obtained from the Ministry of Environment was compared with the predicted land use map obtained from each of the two models. The result from both models showed somewhat similar results. The MSR value obtained from CLUE-S was 0.564, while that from SLEUTH3-R was 0.586. However, when land use map of 2010 was compared with predicted land use map obtained from the two models in same manner, the MSR value obtained from CLUE-S' was 0.500 while that from SLEUTH3-R was decreased to 0.397, an approximately 32.3% decrease from previous value of 2007. Moreover, SLEUTH3-R showed more sensitivity in conversion of urban areas, as compared to other land use types. Therefore, for the prediction of future land use change, CLUE-S model is more reliable than SLEUTH3-R.

• 한국정보통신학회논문지
• /
• 제25권1호
• /
• pp.8-19
• /
• 2021

최근 산업화에 따른 대기오염 문제는 국가와 국민 모두에 큰 관심을 끌고 있다. 국내의 광역(廣域) 대기오염 정보는 국가적으로 공공 데이터를 통해 국민에게 제공하고 있으나, 환경변수가 다른 지역적인 대기오염 정보는 매우 부족한 상황이다. 따라서 본 연구에서는 지역적인 대기오염 현상을 보다 정확하게 분석하고 예측할 수 있는 지역 환경변수 기반의 대기오염 분석 및 예측 시스템을 설계하고 구현한다. 특히 제안한 시스템은 지역적으로 측정된 환경 데이터와 공공 빅데이터를 기반으로 지역의 대기정보를 정확하게 분석하여 제공하고, 인공지능 알고리즘을 통해 미래의 지역 대기정보를 예측하여 제시한다. 나아가 제안된 시스템을 통해 지역적인 대기오염의 발생 원인을 정확하게 파악하여 지역의 대기오염을 예방할 수 있을 것으로 기대할 수 있다.

• 한국물환경학회지
• /
• 제28권1호
• /
• pp.1-9
• /
• 2012

It has been known that Golf course could impose negative impacts on water-ecosystem if pollutant-laden runoff is not treated well. It is important to control non-point source and re-use treated wastewater from the golf course to secure water quality of receiving waterbodies. At golf courses, the rainfall-runoff is affected by various practices to manage grasses. In many hydrological modelings, especially in simple rainfall-runoff modeling, effects on runoff of plant growth and cutting are not considered. In the study, the water erosion prediction project (WEPP), capable of simulating plant growth and various management, was evaluated for its runoff prediction from golf course under grass cutting and irrigation. The %Difference, $R^2$ and the NSE for runoff comparisons were 1.15%, 0.93 and 0.92 for calibration, and 18.12%, 0.82 and 0.88 for validation period, respectively. In grass cutting scenario, grass height was managed to be 18~25 mm. The estimated runoff was decreased by 27%. The difference in estimated total runoff was 11.8% depending on irrigation. As shown in this study, if grass management and irrigation are well-controlled, water quality of downstream areas could be obtained.

• 대기
• /
• 제21권1호
• /
• pp.17-33
• /
• 2011

A dynamical downscaling system for seasonal forecast has been constructed based on a regional climate model, and its predictability was investigated for 10 years' wintertime (December-January-February; DJF) climatology in East Asia. Initial and lateral boundary conditions were obtained from the operational seasonal forecasting data, which are realtime output of the Global Data Assimilation and Prediction System (GDAPS) at Korea Meteorological Administration (KMA). Sea surface temperature was also obtained from the operational forecasts, i.e., KMA El-Nino and Global Sea Surface Temperature Forecast System. In order to determine the better configuration of the regional climate model for East Asian regions, two sensitivity experiments were carried out for one winter season (97/98 DJF): One is for the topography blending and the other is for the cumulus parameterization scheme. After determining the proper configuration, the predictability of the regional forecasting system was validated with respect to 850 hPa temperature and precipitation. The results showed that mean fields error and other verification statistics were generally decreased compared to GDAPS, most evident in 500 hPa geopotential heights. These improved simulation affected season prediction, and then HSS was better 36% and 11% about 850 hPa temperature and precipitation, respectively.

• 대기
• /
• 제16권4호
• /
• pp.259-267
• /
• 2006

In the present study, the neural network (NN) model with cluster analysis method was developed to predict storm surge in the whole Korean coastal regions with special focuses on the regional extension. The model used in this study is NN model for each cluster (CL-NN) with the cluster analysis. In order to find the optimal clustering of the stations, agglomerative method among hierarchical clustering methods was used. Various stations were clustered each other according to the centroid-linkage criterion and the cluster analysis should stop when the distances between merged groups exceed any criterion. Finally the CL-NN can be constructed for predicting storm surge in the cluster regions. To validate model results, predicted sea level value from CL-NN model was compared with that of conventional harmonic analysis (HA) and of the NN model in each region. The forecast values from NN and CL-NN models show more accuracy with observed data than that of HA. Especially the statistics analysis such as RMSE and correlation coefficient shows little differences between CL-NN and NN model results. These results show that cluster analysis and CL-NN model can be applied in the regional storm surge prediction and developed forecast system.

• Ocean and Polar Research
• /
• 제43권3호
• /
• pp.179-192
• /
• 2021

In this study, an integrated wave model from global to coastal scales was developed to improve the operational wave prediction performance of the Korean Meteorological Administration (KMA). In this system, the wave model was upgraded to the WaveWatch III version 6.07 with the improved parameterization of the source term. Considering the increased resolution of the wind input field and the introduction of the high-performance KMA 5^th Supercomputer, the spatial resolution of global and regional wave models has been doubled compared to the operational model. The physical processes and coefficients of the wave model were optimized for the current KMA global atmospheric forecasting system, the Korean Integrated Model (KIM), which is being operated since April 2020. Based on the sensitivity experiment results, the wind-wave growth parameter (β_max) for the global wave model was determined to be 1.33 with the lowest root mean square errors (RMSE). The value of β_max showed the lowest error when applied to regional/coastal wave models for the period of the typhoon season when strong winds occur. Applying the new system to the case of August 2020, the RMSE for the 48-hour significant wave height prediction was reduced by 13.4 to 17.7% compared to the existing KMA operating model. The new integrated wave prediction system plans to replace the KMA operating model after long-term verification.

• 한국지구과학회지
• /
• 제22권3호
• /
• pp.186-194
• /
• 2001

제주 지방 기상청을 대상으로 하는 지역 규모 단시간 수치예보 시스템을 구축하였다. 기상청 본청에서 하루 2회 제공되는 30 km해상도의 수치예보 자료로는 지방 기상청의 예보관들이 우리 나라와 같이 복잡한 지형에서 발생하는 그 지역의 국지 악기상을 파악하기에는 무리가 있다. 지역 규모의 고해상도 수치예보를 위해 LAPS와 MM5를 자료분석과 예보 모델로 이용하였다. LAPS는 양질의 수치예보 초기자료를 생산해 내기 위해 종관 관측 자료뿐만 아니라 위성 및 레이더 등의 비 종관 관측자료도 자료동화에 이용한다. MM5 모델은 16노드의 펜티엄 PC로 구성된 클러스터에서 수행되었으며 이 시스템은 분산병렬 클러스터 컴퓨터로 가격대비 성능이 매우 우수한 미니 슈퍼컴퓨터이다. 자료동화 모델, 수치예보 모델 그리고 PC-클러스터를 종합한 지역 규모 단시간 수치예보 시스템을 한라 단시간 예측 시스템이라 명명하였으며 이 시스템은 현재 제주 지방 기상청에서 독자적으로 운영되고 있다. 기상청 본청에서 제공되는 수치예보 정보로는 탐지할 수 없었던 1999년 7월 9일 제주 지역의 집중호우 사례에 대하여 본 시스템을 검증한 결과 모델이 예측한 강수량이 실제 강수량을 잘 재현하였다. 한라 단시간 예측 시스템은 2000년 4월부터 하루 4회 제주 지방기상청에서 독자적으로 운영되고 있다.

• 환경영향평가
• /
• 제21권1호
• /
• pp.105-117
• /
• 2012

Future climate according to land-use change was simulated by regional climate model. The goal of study was to predict the distribution of meteorological elements using the Weather Research & Forecasting Model (WRF). The KME (Korea Ministry of Environment) medium-category land-use classification was used as dominant vegetation types. Meteorological modeling requires higher and more sophisticated land-use and initialization data. The WRF model simulations with HyTAG land-use indicated certain change in potential vegetation distribution in the future (2086-2088). Compared to the past (1986-1988) distribution, coniferous forest area was decreased in metropolitan and areas with complex terrain. The research shows a possibility to simulate regional climate with high resolution. As a result, the future climate was predicted to $4.5^{\circ}$ which was $0.5^{\circ}$ higher than prediction by Meteorological Administration. To improve future prediction of regional area, regional climate model with HyTAG as well as high resolution initial values such as urban growth and CO2 flux simulation would be desirable.