• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.28 no.1B
• /
• pp.21-32
• /
• 2008

Flood forecasting is a very important tool as one of nonstructural measures for reduction of flood damages in life and property and its accuracy is also an important factor. However, when we apply the Storage Function Model(SFM) which is mainly used for the flood forecasting system in Korea, the determination of the parameters is very important but it is difficult. So, the parameters have been calibrated by using an empirical formulas and judgement of hydrologist. Hence, in this study we perform the sensitivity analysis to understand the parameter characteristics and establish the ranges of parameters of the SFM. Also we do the parameter calibration by using the optimization techniques and objective functions, and evaluate their performances. Especially, we suggest a method to determine proper parameters by using a objective function which can be obtained from flood events. So, we use the suggested method for parameter estimation and compare the estimated parameters with the previously reported parameters. As a result of the application, the estimated parameters by the suggested method showed better than them from the previously reported parameters.

• Journal of Korea Water Resources Association
• /
• v.41 no.9
• /
• pp.929-941
• /
• 2008

The objectives of this study are to propose a system for combined use of a hydrologic and a hydraulic model for urban flood forecast model and to evaluate the system on the $300km^2$ Jungrang urban watershed area, which is relatively large area as an urban watershed and consequently composed of very complex drainage pipes and streams with different land uses. In this study, SWMM for hydrologic model and HEC-RAS for hydraulic model are used and the study area is divided into 25 subbasins. The SWMM model is used for sewer drainage analysis within each subbasin, while HEC-RAS for unstready flow analysis in the channel streams. Also, this study develops a GUI system composed of mean areal precipitation input component, hydrologic runoff analysis component, stream channel routing component, and graphical representation of model output. The proposed system was calibrated for the model parameters and verified for the model applicability by using the observation data. The correlation coefficients between simulated and observed flows at the 2 important locations were ranged on 0.83-0.98, while the coefficients of model efficiency on 0.60-0.92 for the verification periods. This study also provided the possibilities of manhole overflows and channel bank inundation through the calculated water profile of longitudinal and channel sections, respectively. It can be concluded that the proposed system can be used as a surface runoff and channel routing models for urban flood forecast over the large watershed area.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2007.05a
• /
• pp.345-349
• /
• 2007

삽교천 홍수예보시스템은 1999년에 개발되어 현재까지 운영되고 있으나, 개발 이후 유역특성의 변화를 반영한 모형 개선이 이루어지지 않았고, 삽교천 하구둑의 영향을 고려한 모형은 개발되어 있지 않은 실정이다. 이 지역 중에서 특히 천안/아산지역은 급격한 인구증가와 산업화 및 도시화에 의해 면적당 자산의 고도화가 증가하며, 이에 따라 홍수시 피해잠재능은 점점 증가하고 있는 상황이다. 홍수예보 정확도 향상을 위하여 삽교천 유역내 수위관측소 증설에 따른 소유역을 재분할하여 유역특성변화에 따른 수문학적 모형을 재구축하였다. 따라서, 삽교천유역에 신설 및 T/M화된 수문관측소에 대한 소유역 분할과 저류함수법을 이용하기 위한 저류상수를 산정하기위해 기존의 일반 종이지도로 제작된 지형도(1:50,000), 녹지자연도, 지질도, 개략토양도 등을 이용하는 대신 수치지도를 이용하여 저류상수를 산정하였다. 변화된 유역 조건을 가지고 삽교천 유역의 전체 유역 및 하도유출계산을 수행한 후, 측정 결과가 있는 지점의 수문곡선과 비교하여 모형상수가 적절히 산정되었는지 검토하고, 개선된 모형상수를 제시하였다. 또한 홍수예보지점인 원평지점의 선행예보시간을 확보하기 위하여 원평지점 상류의 예당저수지 방류량과 원평지점 수위간의 통계학적모형을 구축하였고, 2시간 이상의 선행예보시간을 확보하였다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.2
• /
• pp.198-203
• /
• 2020

In this paper, we propose an artificial water level prediction system for small river flood prediction. River level prediction can be a measure to reduce flood damage. However, it is difficult to build a flood model in river because of the inherent nature of the river or rainfall that affects river flooding. In general, the downstream water level is affected by the water level at adjacent upstream. Therefore, in this study, we constructed an artificial intelligence model using Recurrent Neural Network(LSTM) that predicts the water level of downstream with the water level of two upstream points. The proposed artificial intelligence system designed a water level meter and built a server using Nodejs. The proposed neural network hardware system can predict the water level every 6 hours in the real river.

• Proceedings of the KIEE Conference
• /
• 2004.07d
• /
• pp.2292-2294
• /
• 2004

At satellite communication system for flood forecasting and warning, VSAT system needs to good performance for aquisition of rainfall and water-level data. But, it has difficult for obtaining good performance because of the rainfall attenuation. Thus, in this paper, we introduced the efficiency plan of the transmission power control for Mini-Hub Station.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.15 no.4
• /
• pp.15-25
• /
• 2012

This study is to develop a web-based real-time agricultural flood management system(RAFMS) for 378 agricultural reservoirs equipped with auto water level gauge stations. The RAFMS was designed to operate linking with Rural Agricultural Water Resource Information System(RAWRIS) which supports data viz. real-time rainfall and water level necessary for RAFMS. The system was constituted to monitor the floods simultaneously at each reservoir by calculating the real-time reservoir inflow from watersheds, water level, and release to downstream. In addition, the system has the prediction function for the flood by applying weather forecasting data from Korea Meteorological Administration(KMA).

• Journal of the Korean Society of Hazard Mitigation
• /
• v.11 no.2
• /
• pp.147-156
• /
• 2011

The purpose of multipurpose reservoir operation in flood season is to reduce the peak flood at a control point by utilizing flood control storage or to minimize flood damage by controlling release and release time. Therefore, the most important thing in reservoir operation for flood season is to determine the optimal release and release time. In this study, goal programming is used for the optimal reservoir operation in flood season. The goal programming minimizes a sum of deviation from the target value using linear programming or nonlinear programming to obtain the optimal alternative for the problem with more than two objectives. To analyze the applicability of goal programming, the historical storm data are utilized. The goal programming is applied to the reservoir system operation as well as single reservoir operation. Chungju reservoir is selected for single reservoir operation and Andong and Imha reservoirs are selected for reservoir system operation. The result of goal programming is compared with that of HEC-5. As a result, it was found that goal programming could maintain the reservoir level within flood control level at the end of a flood season and also maintain flood discharge within a design flood at a control point for each time step. The goal programming operation is different from the real operation in the sense that all inflows are assumed to be given in advance. However, flood at a control point can be reduced by calculating the optimal release and optimal release time using suitable constraints and flood forecasting system.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2022.05a
• /
• pp.147-147
• /
• 2022

In January 2021 heavy flood affected South Kalimantan with causing many casualties. The heavy rainfall is predicted to be generated due to the ENSO (El Nino-Southern Oscillation). The weak La-Nina mode appeared to generate more convective cloud above the warmed ocean and result in extreme rainfall with high anomaly compared to past historical rainfall event. Subsequently, the antecedent soil moisture distribution showed to have an important role in generating the flood response. Saturated flow and infiltration excess mainly contributed to the runoff generation due to the high moisture capacity. The hydro-meteorological processes in this event were deeply analyzed using the coupled atmospheric model of Weather Research and Forecasting (WRF) and the hydrological model extension (WRF-Hydro). The sensitivity analysis of the flood response to the SST anomaly and the soil moisture capacity also compared. Result showed that although SST and soil moisture are the main contributors, soil moisture have more significant contribution to the runoff generation despite of anomaly rainfall occurred. Model performance was validated using the Global Precipitation Measurement (GPM) and Soil Moisture Operational Products System (SMOPS) and performed reasonably well. The model was able to capture the hydro-meteorological process of atmosphere and hydrological feedbacks in the extreme weather event.

• Water for future
• /
• v.25 no.4
• /
• pp.109-119
• /
• 1992

The purpose of this study is to develop real-time streamflow forecasting models in order to manage effectively the flood warning system and water resources during the storm. The stochastic system models of the rainfall-runoff process using in this study are constituted and applied the Recursive Least Square and the Instrumental Variable-Approximate Maximum Likelihood algorithm which can estimate recursively the optimal parameters of the model. Also, in order to improve the performance of streamflow forecasting, initial values of the model parameter and covariance matrix of parameter estimate errors were evaluated by using the observed historical data of the hourly rainfall-runoff, and the accuracy and applicability of the models developed in this study were examined by the analysis of the I-step ahead streamflow forecasts.

• Journal of Wetlands Research
• /
• v.12 no.3
• /
• pp.155-163
• /
• 2010

The flash flood is characterized as flood leading to damage by heavy rainfall occurred in steep slope and impervious area with short duration. Flash flood occurs when rainfall exceeds Flash Flood Guidance(FFG). So, the accurate estimation of FFG will be helpful in flash flood forecasting and warning system. Say, if we can reduce the uncertainty of rainfall-runoff relationship, FFG can be estimated more accurately. However, since the rainfall-runoff models have their own parameter characteristics, the uncertainty of FFG will depend upon the selection of rainfall-runoff model. This study used four rainfall-runoff models of HEC-HMS model, Storage Function model, SSARR model and TANK model for the estimation of models' uncertainties by using Monte Carlo simulation. Then, we derived the confidence limits of rainfall-runoff relationship by four models on 95%-confidence level.