• 한국지반환경공학회 논문집
• /
• 제3권4호
• /
• pp.11-17
• /
• 2002

본 논문의 목적은 섬진다목적댐 유역의 하천을 대상으로 강우시에 단기 수질상태를 예측하기 위하여 병렬다중결선의 계층구조를 갖는 신경망이론을 이용하였다. 본 연구에 적용한 신경망이론의 학습알고리즘으로는 역전파알고리즘을 사용하였으며, 최적모형의 개발을 위해 모멘트법-적응학습율기법을 이용하였다. 하천 수질오염 부하량에 영향을 미치는 요소로서 상류로부터 유입되는 유입량과 수질인자인 BOD, COD, SS를 고려하였다. 섬진다목적댐 유역에 대해 단기 수질을 예측할 수 있는 다층신경망모형을 개발하기 위해 은닉층 노드수와 학습회수에 변화를 주어 각 수질인자별로 4가지씩 총 12개의 모형을 구성하여 학습을 실시하였다. 제안된 신경망모형의 검증을 위해 학습시키지 않은 수질자료를 예측한 결과 양호한 것으로 분석되었고, 하천수계의 단기 수질오염 예측에 활용할 수 있을 것으로 사료되었다.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2008년도 학술발표회 논문집
• /
• pp.2193-2196
• /
• 2008

In this study, the new methodology such as support vector machines neural networks model (SVM-NNM) using the statistical learning theory is introduced to forecast flood stage in Nakdong river, Republic of Korea. The SVM-NNM in hydrologic time series forecasting is relatively new, and it is more problematic in comparison with classification. And, the multilayer perceptron neural networks model (MLP-NNM) is introduced as the reference neural networks model to compare the performance of SVM-NNM. And, for the performances of the neural networks models, they are composed of training, cross validation, and testing data, respectively. From this research, we evaluate the impact of the SVM-NNM and the MLP-NNM for the forecasting of the hydrologic time series in Nakdong river. Furthermore, we can suggest the new methodology to forecast the flood stage and construct the optimal forecasting system in Nakdong river, Republic of Korea.

• 생태와환경
• /
• 제45권2호
• /
• pp.150-157
• /
• 2012

Fish communities in river networks provide significant ecosystem services that will likely decline under future land use and climate change. We developed a model that simulates the consequences to multiple populations of one or more fish species-a meta-community-from multiple stressors across a river network. The model is spatially-explicit and age-structured, with three components: habitat suitability; population dynamics, including species interactions; and movement across a spatial network. Although this model is simple, it can form the basis of fisheries assessments and may be incorporated into an integrated modeling system for watershed management and prediction.

• 한국지리정보학회지
• /
• 제25권2호
• /
• pp.88-99
• /
• 2022

하천망은 하천 관리에 있어서 필수적인 지형특성 중 하나이다. 기존에 현장조사를 통해 구축되었던 하천망은 최근에 원격탐사 자료를 활용하여 효율적으로 구축되기 시작하였다. 교량 등 장애물이 많은 도시 하천망의 경우, 하천 내 장애물 제거에 어려움이 있어 온전한 하천망을 구축한 사례는 드물다. 본 연구는 Sentinel-2 위성영상을 활용하여 도시 내 하천에 존재하는 장애물을 제거하고 경계선이 보전된 온전한 하천망을 자동으로 추출하는 기술을 개발하였다. 우선 Sentinel-2 위성영상의 다중분광 밴드를 활용하여 정규수분지수 영상을 제작하고 수체와 그 외 지역을 구분할 수 있는 이진화 영상을 제작하였다. 그리고 모폴로지 연산을 이진화 영상에 적용하여 장애물이 제거되고 경계선이 보전된 온전한 하천망을 추출하였다. 본 연구에서 개발한 기술을 서울시 한강에 적용한 결과, 경계선은 보존되고 교량 등 장애물이 제거된 온전한 하천망을 추출할 수 있었다.

• 한국방재학회:학술대회논문집
• /
• 한국방재학회 2008년도 정기총회 및 학술발표대회
• /
• pp.669-673
• /
• 2008

In this study, the new methodology which combines Kohonen self-organizing map(KSOM) neural networks model and the conventional neural networks models such as feedforward neural networks model and generalized neural networks model is introduced to forecast flood stage in Nakdong river, Republic of Korea. It is possible to train without output data in KSOM neural networks model. KSOM neural networks model is used to classify the input data before it combines with the conventional neural networks model. Four types of models such as SOM-FFNNM-BP, SOM-GRNNM-GA, FFNNM-BP, and GRNNM-GA are used to train and test performances respectively. From the statistical analysis for training and testing performances, SOM-GRNNM-GA shows the best results compared with the other models such as SOM-FFNNM-BP, FFNNM-BP, and GRNNM-GA and FFNNM-BP shows vice-versa. From this study, we can suggest the new methodology to forecast flood stage and construct flood warning system in river basin.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2007년도 학술발표회 논문집
• /
• pp.1427-1431
• /
• 2007

In this study, the new methodology which combines Kohonen self-organizing map(KSOM) neural networks model and the conventional neural networks models such as feedforward neural networks model and generalized neural networks model is introduced to forecast flood stage in Nakdong river, Republic of Korea. It is possible to train without output data in KSOM neural networks model. KSOM neural networks model is used to classify the input data before it combines with the conventional neural networks model. Four types of models such as SOM-FFNNM-BP, SOM-GRNNM-GA, FFNNM-BP, and GRNNM-GA are used to train and test performances respectively. From the statistical analysis for training and testing performances, SOM-GRNNM-GA shows the best results compared with the other models such as SOM-FFNNM-BP, FFNNM-BP, and GRNNM-GA and FFNNM-BP shows vice-versa. From this study, we can suggest the new methodology to forecast flood stage and construct flood warning system in river basin.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2023년도 학술발표회
• /
• pp.159-159
• /
• 2023

The Amazon River basin is one of the largest basins in the world, and its ecosystem is vital for biodiversity, hydrology, and climate regulation. Thus, understanding the hydrometeorological process is essential to the maintenance of the Amazon River basin. However, it is still tricky to monitor the Amazon River basin because of its size and the low density of the monitoring gauge network. To solve those issues, remote sensing products have been largely used. Yet, those products have some limitations. Therefore, this study aims to do bias corrections to improve the accuracy of Satellite Precipitation Products (SPPs) in the Amazon River basin. We use 331 rainfall stations for the observed data and two daily satellite precipitation gridded datasets (CHIRPS, TRMM). Due to the limitation of the observed data, the period of analysis was set from 1st January 1990 to 31st December 2010. The observed data were interpolated to have the same resolution as the SPPs data using the IDW method. For bias correction, we use convolution neural networks (CNN) combined with an autoencoder architecture (ConvAE). To evaluate the bias correction performance, we used some statistical indicators such as NSE, RMSE, and MAD. Hence, those results can increase the quality of precipitation data in the Amazon River basin, improving its monitoring and management.

• Korean Journal of Hydrosciences
• /
• 제2권
• /
• pp.85-98
• /
• 1991

This paper is focused on the development of the RIVNET1 model, which is a dynamic wave model, for flood analysis in dendritic river networks with arbitrary cross-sections. This model adopted the $-point implicit RDM and utilized a relaxation algorithim in order to solve the governing equations. The double-sweep method was used to reduce the C.P.U. time to solve the matrix system of the model. This model is applied the analyze flood waves of the Ohid river in the U.S.A. and the Keum river in Korea. The results of analysis obtained from this model are compared with those of the DWOPER and observed data.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2016년도 학술발표회
• /
• pp.259-259
• /
• 2016

This study examined Artificial Neurons Networks model (ANNs) for forecast flash discharge at Southern part of Thailand by using rainfall data and discharge data. The Sungai Kolok River Basin has meant the border crossing between Thailand and Malaysia which watershed drains an area lies in Thailand 691.88 square kilometer from over all 2,175 square kilometer. The river originates in mountainous area of Waeng district then flow through Gulf of Thailand at Narathiwat Province, which the river length is approximately 103 kilometers. Almost every year, flooding seems to have increased in frequency and magnitude which is highly non-linear and complicated phenomena. The purpose of this study is to forecast runoff on Sungai Kolok at X.119A gauge station (Sungai Kolok district, Narathiwat province) for 3 days in advance by using Artificial Neural Networks model (ANNs). 3 daily rainfall stations and 2 daily runoff station have been measured by Royal Irrigation Department and Meteorological Department during flood period 2000-2014 were used as input data. In order to check an accuracy of forecasting, forecasted runoff were compared with observed data by pursuing Coefficient of determination ($R^2$). The result of the first day gets the highest accuracy and then decreased in day 2 and day 3, consequently. $R^2$values for first day, second day and third day of runoff forecasting is 0.71, 0.62 and 0.49 respectively. The results confirmed that the ANNs model can be used when the range of collected dataset is short and real-time operated. In conclusion, the ANNs model is suitable to runoff forecasting during flood incident of Sungai Kolok river because it is straightforward model and require with only a few parameters for simulation.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2023년도 학술발표회
• /
• pp.142-142
• /
• 2023

Humans inevitably and continuously produce wastewater in daily life worldwide. To decrease the degradation of river water bodies and aquatic ecosystem therein, humans have built systems at different scales to collect, drain, and treat household-produced wastewater. Particularly, municipal wastewater treatment plants (WWTPs) with centralized controls have played a key role in reducing loads of nutrients in domestic wastewater for the last few decades. Notwithstanding such contributions, impaired rivers regarding water quality and habitat integrity still exist at the whole river basin scale. It is highly attributable to the absence of dilution capacity of receiving streams and/or the accumulation of the pollutant loads along flow paths. To improve the perspective for individual WWTPs assessment, the first crucial step is to achieve systematic understanding on spatial distribution characteristics of all WWTPs together in a given river basin. By taking the initiative, our former study showed spatial hierarchical distributions of WWTPs in three large urbanized river basins in Germany. In this study, we uncover how municipal WWTPs in the contiguous United States are distributed along river networks in a give river basin. The extended spatial scope allows to deal with wide ranges in geomorphological attributes, hydro-climatic conditions, and socio-economic status. Furthermore, we identify the relation of the findings with multiple factors related to human activities, such as the spatial distribution of human settlements, the degree of economy development, and the fraction of communities served by WWTPs. Generalizable patterns found in this study are expected to contribute to establishing viable management plans for recent water-environmental challenges caused by WWTP-discharges to river water bodies.