• Journal of Korean Society on Water Environment
• /
• v.25 no.4
• /
• pp.573-579
• /
• 2009

The performance of Rainfall-Runoff Forecasting System (RRFS), the watershed management model for the Geum river basin, is evaluated based on the agreement between the simulated and observed hydrographs and the behavioral characteristics of the flow-duration curves. As a result, the simulated hydrographs are well agreed with the observed ones except high flow discharges. It is inferred that most of the errors in the simulated hydrographs are due to the misestimation of agricultural water use in $2^{nd}$ quarter and the discrepancy of the peak discharges in $3^{rd}$ quarter. It is however judged that RRFS would give the reliable runoff hydrographs from the point of view of continuous model application. And simulated flow-duration curves and flow-duration coefficients are also similar to the observed ones except flood flow region. From the above result it is confirmed that the construction of Yongdam dam improves the state of flow-duration curve at the Gongjoo station.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2021.06a
• /
• pp.373-373
• /
• 2021

Rainfall forecasting is an important issue that is applied in many areas, such as agriculture, flood warning, and water resources management. In this context, this study proposed a statistical and machine learning-based forecasting model for monthly rainfall. The Bayesian Gaussian process was chosen to optimize the hyperparameters of the Stacked Long Short-term memory (SLSTM) model. The proposed SLSTM model was applied for predicting monthly precipitation of Seoul station, South Korea. Data were retrieved from the Korea Meteorological Administration (KMA) in the period between 1960 and 2019. Four schemes were examined in this study: (i) prediction with only rainfall; (ii) with deseasonalized rainfall; (iii) with rainfall and minimum temperature; (iv) with deseasonalized rainfall and minimum temperature. The error of predicted rainfall based on the root mean squared error (RMSE), 16-17 mm, is relatively small compared with the average monthly rainfall at Seoul station is 117mm. The results showed scheme (iv) gives the best prediction result. Therefore, this approach is more straightforward than the hydrological and hydraulic models, which request much more input data. The result indicated that a deep learning network could be applied successfully in the hydrology field. Overall, the proposed method is promising, given a good solution for rainfall prediction.

• Journal of Korea Water Resources Association
• /
• v.51 no.12
• /
• pp.1207-1216
• /
• 2018

Discharge or water level predictions at tidally affected river reaches are currently still a great challenge in hydrological practices. This research aims to predict water level of the tide dominated site, Jamsu bridge in the Han River downstream. Physics-based hydrodynamic approaches are sometimes not applicable for water level prediction in such a tidal river due to uncertainty sources like rainfall forecasting data. In this study, TensorFlow deep learning framework was used to build a deep neural network based LSTM model and its applications. The LSTM model was trained based on 3 data sets having 10-min temporal resolution: Paldang dam release, Jamsu bridge water level, predicted tidal level for 6 years (2011~2016) and then predict the water level time series given the six lead times: 1, 3, 6, 9, 12, 24 hours. The optimal hyper-parameters of LSTM model were set up as follows: 6 hidden layers number, 0.01 learning rate, 3000 iterations. In addition, we changed the key parameter of LSTM model, sequence length, ranging from 1 to 6 hours to test its affect to prediction results. The LSTM model with the 1 hr sequence length led to the best performing prediction results for the all cases. In particular, it resulted in very accurate prediction: RMSE (0.065 cm) and NSE (0.99) for the 1 hr lead time prediction case. However, as the lead time became longer, the RMSE increased from 0.08 m (1 hr lead time) to 0.28 m (24 hrs lead time) and the NSE decreased from 0.99 (1 hr lead time) to 0.74 (24 hrs lead time), respectively.

• Journal of Environmental Science International
• /
• v.13 no.7
• /
• pp.611-618
• /
• 2004

The Neural Network Models which mathematically interpret human thought processes were applied to resolve the uncertainty of model parameters and to increase the model's output for the streamflow forecast model. In order to test and verify the flood discharge forecast model eight flood events observed at Kumho station located on the midstream of Kumho river were chosen. Six events of them were used as test data and two events for verification. In order to make an analysis the Levengerg-Marquart method was used to estimate the best parameter for the Neural Network model. The structure of the model was composed of five types of models by varying the number of hidden layers and the number of nodes of hidden layers. Moreover, a logarithmic-sigmoid varying function was used in first and second hidden layers, and a linear function was used for the output. As a result of applying Neural Networks models for the five models, the N10-6model was considered suitable when there is one hidden layer, and the Nl0-9-5model when there are two hidden layers. In addition, when all the Neural Network models were reviewed, the Nl0-9-5model, which has two hidden layers, gave the most preferable results in an actual hydro-event.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2017.05a
• /
• pp.379-379
• /
• 2017

과거에는 하천 범람으로 인한 홍수피해가 많았으나 최근에는 도시화로 인한 불투수면적의 증가로 홍수도달시간의 단축 및 노면수의 배수불량으로 인한 내수 홍수피해가 많아졌다. 이러한 변화는 도시하천의 홍수예보에 밀접한 관련이 있으며 관련된 분석 모형 및 연계방안 또한 매우 중요하게 되었다. 일반적으로 하천에 대한 유출해석 모형으로 HEC-RAS((Hydrologic Engineering Center-River Analysis System)가 주로 사용되고 있으나 현재와 같이 도심지 하천에서는 내배수의 특성을 고려한 SWMM(Storm Water Management Model)을 사용한다. 또는 이 두모형의 연계를 통해 유출해석을 진행하기도 한다. 최근 HEC-RAS와 SWMM모형이 최신 버전을 공개하였다. HEC-RAS의 경우 2016년 9월 5.0.3버전을 출시하며 1D뿐만 아닌 2D의 모의도 가능하도록 기능을 개선하였으며 SWMM의 경우 2016년 09월 07일 5.1.011버젼이 공개되었다. 본 연구에서는 공개된 최신 모형을 도림천 지역에 적용하여 도림천 지역에 적합한 모형 및 연계 방법을 찾아보려 한다. 이를 통해 최적의 도시홍수예보 시스템을 구성하기 위한 모형 및 연계방안의 조사와 가장 합리적인 도시홍수 시스템의 구성방안을 제시하고자 한다.

• Journal of Wetlands Research
• /
• v.23 no.2
• /
• pp.144-153
• /
• 2021

In recently, the green algae bloom is one of the most severe challenges. The seven days prior prediction is in operation to issues the water quality warning, but it also needs a longer time of prediction to take preemptive measures. The objective of the study is to establish a method to conduct a 3-month prior prediction of Chl-a concentration in the Daechong Lake and tested its applicability as a supplementary of current water quality warning. The historical record of water quality in the Daechong Lake and seasonal forecasting of ECMWF were obtained, and its time-series characteristics were analyzed. The Chl-a forecasting model was established using a correlation between Chl-a concentration and meteorological factor and NARX model, and its efficiency was compared.

• Journal of Korea Water Resources Association
• /
• v.46 no.6
• /
• pp.655-666
• /
• 2013

Lately, intensity and frequency of natural disasters such as flood are increasing because of abnormal climate. Casualties and property damages due to large-scale floods such as Typhoon Rusa in 2002 and Typhoon Maemi in 2003 rapidly increased, and these show the limits of the existing disaster prevention measures and flood forecasting systems regarding irregular climate changes. In order to efficiently respond to extraordinary flood, it is important to provide effective countermeasures through an inundation model that can accurately simulate flood inundation patterns. However, the existing flood inundation analysis model has problems such as excessive take of analysis time and accuracy of the analyzed results. Therefore, this study conducted a flood inundation analysis by using the Gerris flow solver that uses quadtree grid, targeting the Baeksan Levee in the Nakdong River Basin that collapsed because of a concentrated torrential rainfall in August, 2002. Through comparisons with the FLUMEN model that uses unstructured grid among the existing flood inundation models and the actual flooded areas, it determined the applicability and efficiency of the quadtree grid-based flood inundation model of the Gerris flow solver.

• Journal of Korea Water Resources Association
• /
• v.53 no.12
• /
• pp.1059-1070
• /
• 2020

River cross-section measurement data is one of the most important input data in research related to hydraulic and hydrological modeling, such as flow calculation and flood forecasting warning methods for river management. However, the acquisition of accurate and continuous cross-section data of rivers leading to irregular geometric structure has significant limitations in terms of time and cost. In this regard, a primary objective of this study is to develop a methodology that is able to measure the spatial distribution of continuous river characteristics by minimizing the input of time, cost, and manpower. Therefore, in this study, we tried to examine the possibility and accuracy of continuous cross-section estimation by estimating the water depth for each cross-section through multiple linear regression analysis using RGB-based aerial images and actual data. As a result of comparing with the actual data, it was confirmed that the depth can be accurately estimated within about 2 m of water depth, which can capture spatially heterogeneous relationships, and this is expected to contribute to accurate and continuous river cross-section acquisition.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2012.05a
• /
• pp.301-304
• /
• 2012

홍수예경보는 강우로 인하여 발생되는 홍수의 규모와 시간을 가능한 한 정확하고 빨리 예측하여 홍수에 대비할 수 있도록 유관기관 및 지역주민에게 사전에 홍수에 관한 정보 즉 예측되는 수위와 시간을 제공함으로써 홍수로부터의 피해를 최소화하는 것이다. 이와 같은 목적을 성공적으로 완수하기 위해서는 홍수시 급변하는 하천유량에 영향을 미치는 모든 수문학적 기상학적 자료를 신속 정확하게 수집할 수 있는 관측 시스템의 구축 뿐 아니라 이들 수집된 자료를 이용하여 실시간 홍수추적을 할 수 있는 효율적인 유출량 계산모형이 조화를 이룰 때 가능하다. 이에 본 연구에서는 중 소하천에서 홍수예경보를 위한 지능형 U-River 시스템의 실시간 모니터링 기술을 조사하고 하천수위를 이용한 예측시스템에 대해 연구하였다. 기존의 홍수예경보의 문제점을 해결하기 위해 간단한 입력자료만으로 홍수예측이 가능한 인공지능 기반의 신경망 모형을 이용 하였으며, 예측 모형의 효율성과 적용성을 높이기 위해 유사한 수문 사상을 가지는 상 하류간 입력 자료를 동시에 사용하였다. 또한 하천수위를 이용한 모델의 수행은 각 지점별 훈련성과를 토대로 최적의 은닉층 노드수를 선발하여 실시간 수위예측에 활용하였으며 수치적 기준을 적용하여 실측 수위와 모형에 의해 예측된 수위를 이용하여 평가하였다.

• Magazine of the Korean Society of Agricultural Engineers
• /
• v.39 no.6
• /
• pp.54-66
• /
• 1997

The objectives of this study is to introduce and apply neural network theory to real hydrologic systems for stochastic nonlinear predicting of daily runoff discharge in the river catchment. Back propagation algorithm of neural network model is applied for the estimation of daily stochastic runoff discharge using historical daily rainfall and observed runoff discharge. For the fitness and efficiency analysis of models, the statistical analysis is carried out between observed discharge and predicted discharge in the chosen runoff periods. As the result of statistical analysis, method 3 which has much processing elements of input layer is more prominent model than other models(method 1, method 2) in this study.Therefore, on the basis of this study, further research activities are needed for the development of neural network algorithm for the flood prediction including real-time forecasting and for the optimal operation system of dams and so forth.