• Journal of Korean Society of Water and Wastewater
• /
• v.36 no.4
• /
• pp.239-248
• /
• 2022

The prediction of algal bloom is an important field of study in algal bloom management, and chlorophyll-a concentration(Chl-a) is commonly used to represent the status of algal bloom. In, recent years advanced machine learning algorithms are increasingly used for the prediction of algal bloom. In this study, XGBoost(XGB), an ensemble machine learning algorithm, was used to develop a model to predict Chl-a in a reservoir. The daily observation of water quality data and climate data was used for the training and testing of the model. In the first step of the study, the input variables were clustered into two groups(low and high value groups) based on the observed value of water temperature(TEMP), total organic carbon concentration(TOC), total nitrogen concentration(TN) and total phosphorus concentration(TP). For each of the four water quality items, two XGB models were developed using only the data in each clustered group(Model 1). The results were compared to the prediction of an XGB model developed by using the entire data before clustering(Model 2). The model performance was evaluated using three indices including root mean squared error-observation standard deviation ratio(RSR). The model performance was improved using Model 1 for TEMP, TN, TP as the RSR of each model was 0.503, 0.477 and 0.493, respectively, while the RSR of Model 2 was 0.521. On the other hand, Model 2 shows better performance than Model 1 for TOC, where the RSR was 0.532. Explainable artificial intelligence(XAI) is an ongoing field of research in machine learning study. Shapley value analysis, a novel XAI algorithm, was also used for the quantitative interpretation of the XGB model performance developed in this study.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2005.09a
• /
• pp.157-159
• /
• 2005

• Proceedings of the Korea Water Resources Association Conference
• /
• 2006.05a
• /
• pp.1945-1949
• /
• 2006

A storm-induced coastal inundation model (SICIM) is presented to simulate the flood event during typhoon passage that often results in significant rise in sea-level heights especially in the upstream region of the basin. The SICIM is a GIS-based distributed hydrodynamic model, both storm surge and storm water inundations are taken into account. The spatial and temporal distribution of the storm water level and flux are calculated. The model was applied to Jeju Island since it has an isolated watershed that is easy to handle as a first step of model application. Another reason is that it is surrounded by coastal area exposed to storm surge inundation. The model is still advancing and will be the framework of a predictive early inundation warning system.

• Journal of Environmental Science International
• /
• v.5 no.3
• /
• pp.377-385
• /
• 1996

This study aims at the estimation of a river management flow in urban basin analyzing Sinchun basin to be the tributary of Kumho river basin. The river management flow has to satisfy a low flow as natural flow and an environmental preservation flow estimated by a dilution flow to satisfy a target water quality in drought flow. Therefore for the estimation of a river management flow in Sinchun in this study, first Tank model as a basin runoff model estimates a low flow, a drought flow from a flow duration curve in Sinchun, second QUAL2E model as water quality model simulates water quality in Sinchun and estimates environmental preservation flow to satisfy a target water qua%its, BOD 8 mg/l by a dilution flow derived from Kumho river, Nakdong river and around water. And the river management flow is estimated by addition of a use flow and a loss flow to more flow between a low-flow and an environmental preservation flow.

• Proceedings of the Korean Society of Agricultural Engineers Conference
• /
• 2000.10a
• /
• pp.578-583
• /
• 2000

In order to operate water quality model for a lake or a channel, user should examine the all kinds of parameters and should know how them react to the model for calculating the pollution which are happened from the watershed or are reacted in the water. The aim of this study is analyzing the characteristics of parameters which are used by a water quality model (CE-QUAL-W2), so that we are trying to find out how them to react to the model for calculating the many kinds of pollution.

• Proceedings of the Korea Committee for Ocean Resources and Engineering Conference
• /
• 2000.10a
• /
• pp.246-252
• /
• 2000

The water quality in Yeongil Bay is getting worse due to the sewage and the waste water from the surrounding industrial complex The study aims to simulate the current system that is necessary to built ecosystem model for the optium water quality control and clarify the correlation of current system characteristics with water quality in Yongil Bay. To clarify the characteristics of coastal water movement system and verify the applicability of the 3-D model, the current system was simulated using 3-D baroclinic model considered tidal current and density effects. As the results of numerical experiments, it is proved the 3-D model is the most appliable on the Yongil Bay where current flows slowly and the flow direction is varied by depths. From the results of simulation considered tidal current only, It am be clearly said the water in Yongil Bay flows in through the surface layer and flows out through the bottom layer. And the fresh water from the Hyongsan river and the heated discharge from POSCO have little effect on the current structure in Yonggil Bay, but have and important effect upon the density structure by diffusion of heat and salt. And the water quality distribution is closely related with the current structure characteristics as well as the tidal residual current system.

• Journal of Korean Society on Water Environment
• /
• v.27 no.5
• /
• pp.587-596
• /
• 2011

The interest in hydrological modeling has increased significantly recently due to the necessity of watershed management, specifically in regards to lumped models, which are being prosperously utilized because of their relatively uncomplicated algorithms which require less simulation time. However, lumped models require empirical coefficients for hydrological analyses, which do not take into consideration the heterogeneity of site-specific characteristics. To overcome such obstacles, a distributed model was offered as an alternative and the number of researches related to watershed management and distributed models has been steadily increasing in the recent years. Thus, in this study, the feasibility of a grid-based rainfall-runoff model was reviewed using the flood runoff process in the Han River basin, including the ChungjuDam, HoengseongDam and SoyangDam watersheds. Hydrological parameters based on GIS/RS were extracted from basic GIS data such as DEM, land cover, soil map and rainfall depth. The accuracy of the runoff analysis for the model application was evaluated using EFF, NRMSE and QER. The calculation results showed that there was a good agreement with the observed data. Besides the ungauged spatial characteristics in the SoyangDam watershed, EFF showed a good result of 0.859.

• Journal of Korean Society on Water Environment
• /
• v.25 no.5
• /
• pp.651-660
• /
• 2009

Detailed flow analysis in river is essential to increase the accuracy of water quality simulation since flow variation depends on many factors such as cross sections, channel slopes, and bed materials. In the QUAL2E stream water quality simulation model, the hydraulic coefficients are assigned to the reach that is collection of computational element using the hydraulic coefficient. This study developed a module that can incorporate the results of non-uniform flow analysis and assign such information to each individual element. Model application focused on the upstream of the Hoengseong reservoir including the reservoir where significant flow change is expected. Comparing with original QUAL2E model the developed module improved the result of water quality simulation without considering the relation of flow velocity and flow depth in terms of flow rates.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2007.05a
• /
• pp.1910-1914
• /
• 2007

This paper employs a relatively new technique of Artificial Neural Network (ANN) to forecast water demand of Daegu city. The ANN model used in this study is a single hidden layer hierarchy model. About seventeen sets of historical water demand records and the values of their socioeconomic impact factors are used to train the model. Also other regression and time serious models are investigated for comparison purpose. The results present the ANN model can better perform the issue of urban water demand forecasting, and obtain the correlation coefficient of $R^2$ with a value of 0.987 and the relative difference less than 4.4% for this study.

• Proceedings of the Korea Water Resources Association Conference
• /
• 1988.07a
• /
• pp.85-92
• /
• 1988

A model for the transient one-dimensional moisture movement in the saturated-unsaturated zone using a finite difference method is developed. Hysteresis in the soil water retention is incorporated. The model considers layered geologic formations. Monte Carlo simulation, together with the nearest neighbor model is used. Outputs of the model include pressure head, water content, and the water table elevation. Two Monte Carlo simulations of 100 realizations each are made for a 12-day simulation period with different input values. The simulation results show that the S.D. of the outputs increases with an increase in the input, the S.D. of the log K$$. The model is applied to predict a long term water table fluctuation, and the predicted water table agress well with the observed one.