• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1996.10a
• /
• pp.253-256
• /
• 1996

Rainfall is one of the major and complicated elements of hydrologic system. Accurate prediction of rainfall is very important to mitigate storm damage. The neural network is a good model to be applied for the classification problem, large combinatorial optimization and nonlinear mapping. In this dissertation, rainfall predictions by the neural network theory were presented. A multi-layer neural network was constructed. The network learned continuous-valued input and output data. The network was used to predict rainfall. The online, multivariate, short term rainfall prediction is possible by means of the developed model. A multidimensional rainfall generation model is applied to Seoul metropolitan area in order to generate the 10-minute rainfall. Application of neural network to the generated rainfall shows good prediction. Also application of neural network to 1-hour real data in Seoul metropolitan area shows slightly good predictions.

• Bulletin of the Korean Space Science Society
• /
• 2006.10a
• /
• pp.47-47
• /
• 2006

• Magazine of the Korean Society of Agricultural Engineers
• /
• v.45 no.7
• /
• pp.83-93
• /
• 2003

This research compared the observed and model predicted results that include; runoff, sediment yield, and nutrient losses from a 2.71 ha cattle grazing pasture field in North Alabama. Application of water quality computer simulation models can inexpensively and quickly assess the impact of pasture management practices on water quality. AGNPS single storm based model was applied to the three pasture species; Bermudagrass, fescue, and Ryegrass. While comparing model predicted results with observed data, it showed that model can reasonably predict the runoff, sediment yield and nutrient losses from the watershed. Over-prediction and under-prediction by the model occurred during very high and low rainfall events, respectively. The study concluded that AGNPS model can be reasonably applied to assess the impacts of pasture management practices and chicken litter application on water quality.

• Korean Journal of Hydrosciences
• /
• v.3
• /
• pp.31-44
• /
• 1992

Precise run-off forecasting depends on the ability to predict quantitative rainfall intensity. The purpose of this study is to develop a stochastic model for the shori-term rainfall prediction. It is required for the model to predict rainfall intensities at all the telemetered rain-gauge locations simultaneously. All the model parameters, which are used in this work ; velocity and direction of storm movement, radial spectrum, and dimensionless time distribution of rainfall, are the results of the previous study. We formulated the model and operated it, so that in this study was analyzed particulary the influence of 4 dimensionless time distributions on the prediction and the influence of the model on run-off.

• Korean Journal of Hydrosciences
• /
• v.2
• /
• pp.1-12
• /
• 1991

To estimate the peak discharge from the small rural watersheds, 53 storm events of seven small watersheds from 1972 to 1989 were selected and were analyzed by the multiple regression technique. The peak values by the new prediction method developed in this study were also compared to the real data of Banwol Basins and the estimated peaks of the several watersheds which were analyzed by the Korean engineering companies. These values were also compared to the results from the other method, i.e. the Rational Method, the Kajiyama Method, the Nakayasu Unit Hydrograph companies. Through Method, the Area Routing Method, etc., which are favored by the Korean engineering companies. Through these comparisons, it is proved that the proposed method may be used for day-to-day use without any problem. However, there should be some modifications and improvements as more data are available in the future.

• Water for future
• /
• v.24 no.3
• /
• pp.83-94
• /
• 1991

In this paper a discrete extended Kalman filter for the tidal prediction has been developed. The filter is based on a set of difference equations derived from the one dimensional shallow water equations using the finite difference scheme proposed by Lax-Wendroff. The filter gives estimates of the water level and water velocity, together with the parameters in the model which essentially have a random character, e.g. bottom friction and wind stress. The estimates are propagated and updated by the filter when the physical circumstances change. The Kalman-filter is applied to field data gathered in the coastal area alon the West Sea and it is shown that the filter gives satisfactory results in forecasting the waterlevels during storm surge periods.

• Water for future
• /
• v.23 no.3
• /
• pp.341-350
• /
• 1990

Precise run-off forecasting depends on the ability to predict quantitative rainfall intensity. This study suggests a stochastic model for 1 hour order rainfall prediction. The model simultaneously predicts rainfall intensity at all telemetered rain-gauge locations. All model parameters, velocity and direction of storm movement, radial spectrum, dimensionless time distribution of rainfall, are estimated from telemetered and historical data for the basin being predicted. Also the estimated parameters are based on the previous study. The results are the influence of dimensionless time distributions on the prediction and the model on run-off.

• Environmental Engineering Research
• /
• v.10 no.2
• /
• pp.62-70
• /
• 2005

Pollution loading of Combined Sewer Overflows (CSOs) is frequently over the capacity of a wastewater treatment plant (WWTP) receiving the water. The objectives of this study are to investigate water quality of CSOs in Anmyun-ueup, Tean province and to apply Storm Water Management Model to predict flow rate and water quality of the CSOs. The capacity of a local WWTP was also estimated according to rainfall duration and intensity. Eleven water quality parameters were analyzed to characterize overflows. SWMM model was applied to predict the flow rate and pollutant load of CSOs during rain event. Overall, profile of the flow and pollutant load predicted by the model well followed the observed data. Based on model prediction and observed data, CSOs frequently occurs in the study area, even with light precipitation or short rainfall duration. Model analysis also indicated that the local WWTP’s capacity was short to cover the CSOs.

• Journal of the Korean Society of Hazard Mitigation
• /
• v.9 no.5
• /
• pp.131-138
• /
• 2009

In this paper, a dynamic warning system to forecast inland flooding associated with typhoons and storms is described. The system is used operationally during the typhoon season to anticipate the potential impact such as inland flooding on the coastal zone of interest. The system has been developed for the use of the public and emergency management officials. Simple typhoon models for quick prediction of wind fields are implemented in a user-friendly way by using a Graphical User Interface (GUI) of MATLAB. The main program for simulating tides, depth-averaged tidal currents, wind-driven surges and currents was also vectorized for the fast performance by MATLAB. By pushing buttons and clicking the typhoon paths, the user is able to obtain real-time water level fluctuation of specific points and the flooding zone. This system would guide local officials to make systematic use of threat information possible. However, the model results are sensitive to typhoon path, and it is yet difficult to provide accurate information to local emergency managers.

• Journal of the Korean earth science society
• /
• v.27 no.6
• /
• pp.653-658
• /
• 2006

The seasonal predictability of typhoon activity over the western North Pacific is investigated using an atmospheric general circulation model GCPS. A ten-member ensemble with different initial conditions is integrated for five months using observed sea surface temperature data for each year from 1979 to 2003. It is shown that the monthly variation of occurrence frequency of simulated tropical storms and the distribution of tropical storm genesis location are similar to those of observed tropical storms, but the model is unable to reliably predict the interannual variation of the occurrence frequency of tropical storms. This is largely because the observed relationship between tropical storm occurrence frequency and ENSO is different from the simulated one. Unlike the observation, in which the tropical storm occurrence frequency has no relation to ENSO, the model has a tendency to generate more (less) tropical storms than normal during El Nino (La Nina). On the other hand, the interannual variation of the mean longitude of tropical storms that shows a close connection with ENSO in both observations and simulations is simulated similar to the observation.